 

 

(A LIBRARY
M Q Michigan State
N7 University

 

 

This is to certify that the
thesis entitled

MATERIAL PROPERTY DOCUMENTATION AND
FRACTURE ANALYSES OF THE DEVELOPING SKULL

presented by

TIMOTHY G. BAUMER

has been accepted towards fulﬁllment
of the requirements for the

MS. degree in Engineering Mechanics

 

 

MWCAWL/

ﬂ Mgﬂar Professor's Signature

M

Date

MSU is an Alﬁrmatlvo ActiorVEquaI Opportunity Employer

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5108 K:IProj/Acc&Pres/CIRCIDateDue.indd

 

MATERIAL PROPERTY DOCUMENTATION AND FRACTURE
ANALYSES OF THE DEVELOPING SKULL

By

Timothy G. Baumer

A THESIS

Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of

MASTER OF SCIENCE

Engineering Mechanics

2009

ABSTRACT

MATERIAL PROPERTY DOCUMENTATION AND FRACTURE
ANALYSES OF THE DEVELOPING SKULL

By
Timothy G. Baumer
Skull fracture interpretation is a necessary component of forensic analysis and
contributes to the determination of the cause and manner of injury or death. The research
presented in this thesis makes use of an in situ porcine model and ﬁnite element analysis
to examine the responses of the developing skull to blunt force impulse and quasi-static
crushing trauma. Chapter 2 documents age effects on the mechanical properties of
parietal bone and coronal suture in porcine infants and correlates the bending properties
of the bone to existing human infant data. Chapter 3 documents skull ﬁacture on infant
porcine specimens with respect to age and interface in an effort to identify ﬁ'acture
characteristics for a developing skull under known conditions. Chapter 4 uses a
simpliﬁed cranial model and the ﬁnite element method to generate representative
patterns of principal stress and strain directions developed during quasi-static crushing
injuries to the developing skull. The theoretical results were compared to four clinical
cases of known childhood fatalities from crushing head injuries inﬂicted by vehicle tires.
Chapter 5 uses a simpliﬁed porcine cranial model and the ﬁnite element method to
predict experimentally documented fracture patterns developed during impact loading to
the parietal bone. The material properties of the cranial model were taken from the
previously obtained results from Chapter 2. The experimentally inﬂicted fractures to the
cranium ﬁ'om Chapter 3 were used for comparison with the theoretically developed

principal stress and strain directions.

ACKNOWLEDGMENTS

I would like to thank my advisor and mentor, Dr. Roger Haut, for all his support and
guidance over the past 4 years. I am also grateful to Dr. Todd Fenton for all his support
and willingness to teach anthropology to an engineer. I would also like to thank all the
members of the 4th ﬂoor of East Fee Hall for their ﬁiendship and support. I am grateful
to Jean Atkinson, Jane Walsh, and Cliff Beckett for their continual technical support.
Finally, I would like to acknowledge the contributions of Mr. Ed Reed and Ms. Star

Lewis for supplying and collecting the porcine specimens.

iii

TABLE OF CONTENTS

LIST OF TABLES .......................................................................................................... vi
LIST OF FIGURES ............................................................................................... i ......... vii
RESEARCH PUBLICATIONS ....................................................................................... x
CHAPTER 1: Characteristics of the Developing Skull - An Introduction ...................... 1
CHAPTER 2: Age-Dependent Mechanical Properties of the Infant Porcine
Parietal Bone and a Correlation to the Human ......................................................... 10
Abstract .............................................................................................................. 10
Introduction ........................................................................................................ 12
Materials and Methods ....................................................................................... 14
Results ................................................................................................................ 18
Discussion .......................................................................................................... 23
References .......................................................................................................... 28

CHAPTER 3: Age-Dependent Fracture Characteristics of Rigid and Compliant

Surface Impacts on the Infant Skull — A Porcine Model .......................................... 30
Abstract .............................................................................................................. 30
Intrdduction ........................................................................................................ 32
Materials and Methods ....................................................................................... 34
Results ................................................................................................................ 38
Discussion .......................................................................................................... 42
References .......................................................................................................... 47

CHAPTER 4: Characteristics and Prediction of Cranial Crush Injuries in Children ..... 50
Abstract .............................................................................................................. 50
Introduction ........................................................................................................ 52
Materials and Methods ....................................................................................... 54
Results ................................................................................................................ 56
Discussion .......................................................................................................... 61
References .......................................................................................................... 65

CHAPTER 5: Finite Element Modeling of Rigid and Compliant Surface Impacts

to the Pediatric Porcine Parietal Bone ...................................................................... 66
Abstract .................................................................. 66
Introduction ........................................................................................................ 68
Materials and Methods ....................................................................................... 70
Results ................................................................................................................ 72
Discussion .......................................................................................................... 74
References .......................................................................................................... 78

CHAPTER 6: Conclusions and Recommendations for Future Work ............................ 80

iv

APPENDD( A ................................................................................................................ 83

APPENDIX B ................................................................................................................. 94

LIST OF TABLES

Table 5.1. (Average) properties used in development of the FE model and impact
conditions. .................................................................................................... 71

Table A. 1. Raw data collected from 4-point bending tests of bone-suture-bone
specimens ..................................................................................................... 84

Table A2. Raw data collected ﬁ'om 4-point bending tests of perpendicular bone
specimens ..................................................................................................... 86

Table A3. Raw data collected from 4-point bending tests of parallel bone
specimens ..................................................................................................... 88

Table A4 Raw data collected from 4-point bending tests of bone-suture-bone
specimens NOT resulting in suture failure ................................................... 90

Table A5. Raw data collected ﬁ'om 4-point bending tests for multiple bone-
suture-bone specimens taken from the same cranium .................................. 91

Table A6. Porosity data from stained cross-sections of perpendicular bone

specimens ..................................................................................................... 92
Table A.7. Porosity data ﬁ'om stained cross-sections of parallel bone specimens ......... 93
Table B.1. Raw data collected ﬁom rigid interface impacts .......................................... 95
Table 3.2. Skull damage measurements for rigid interface impacts .............................. 97

Table 83. Raw data collected from compliant (1.1 MPa Hexcel) interface impacts 98

Table 3.4. Skull damage measurements for compliant interface impacts ................... 100

LIST OF FIGURES

Figure 1.1. Diagram of a newborn skull ........................................................................... 2
Figure 2.1. Top view of cranium with locations of the beam specimens ....................... 15
Figure 2.2. Diagram of testing ﬁxture ............................................................................ 16

Figure 2.3. Representative stained cross-section of a beam from a 14 day
specimen showing separation of compact and porous layers. Voids in
the bone matrix were colored black. The box represents the analyzed
area of a test specimen .................................................................................. 18

Figure 2.4. Recorded values of stiffness for (a) both orientations of bone and (b)
bone-suture-bone specimens ﬂour 3 to 21 days. *Signiﬁcantly
different means ............................................................................................. 19

Figure 2.5. Linear regressions of ultimate stress for bone and bone-suture—bone
specimens against age ...... . .......... 20

 

Figure 2.6. Ultimate strain for bone and bone-suture-bone specimens.
*Signiﬁcantly different means ..................................................................... 20

Figure 2.7. Bending modulus for bone and bone-suture-bone specimens versus
age. *Signiﬁcantly different means .............................................................. 21

Figure 2.8. Strain energy of bone and bone-suture bone specimens versus age.
*Signiﬁcantly different means ..................................................................... 22

Figure 2.9. Linear regression analyses of both orientations of infant porcine bone
specimens against age. Both orientations displayed signiﬁcant
dependencies on age ..................................................................................... 22

Figure 2.10. Bending rigidity of human and porcine specimens versus age
(Human - months; porcine — days). The human data was based on data
ﬁom Coats and Margulies (2006) and unpublished data from Li et al

(2007) ........................................................................................................... 24
Figure 3.1. Skull positioning and set-up showing rigid impact head ............................. 35
Figure 3.2. Drop test ﬁxture set-up ................................................................................ 36
Figure 3.3. Overlay of force-time plots showing typical characteristics of an

impact that caused fracture and one that did not .......................................... 38
Figure 3.4. Peakiforce during impact for rigid and compliant interfaces ....................... 39

vii

Figure 3.5. Contact area recorded on pressure sensitive ﬁhn during impacts for
rigid and compliant interfaces ...................................................................... 39

Figure 3.6. Representative fracture sites for (a) rigid and (b) compliant interface
impacts (ﬁve-day-old specimens shown). Fracture sites initiated at
bone-suture boundaries. Compliant interface impact showed increased
fracture lengths at similar initiation sites as the rigid interface impacts
as well as diastatic fracture of coronal suture ............................................... 40

Figure 3.7. Frequency of diastatic fractures caused by both rigid and compliant
interfaces versus age ..................................................................................... 41

Figure 3.8. (Average) Fracture length generated by rigid and compliant impacts at
the same impact energy ................................................................................ 42

 

Figure 4.1. Simpliﬁed cranial model showing modeled landmarks of the
basicranium. View of endocranial surface with vault removed ................... 55

Figure 4.2. Diagrams of fractures due to quasi-static crushing forces to skulls aged
(a) 5 years -— Case 1, (b) 6 years — Case 2, (c) 3 years - Case 3, and (d)
1.5 years — Case 4. White cranial margins represent contact of applied
forces (ground and tire) ................................................................................ 58

Figure 4.3. (Case 1) Crushing injury to the skull of a 5-year-old. Diagrams of (a)
clinically observed ﬁactures and (b) an overlay of the highest 20% of
principal tensile stresses produced from a FE model. White cranial
margins represent contact of applied forces (ground and tire) ..................... 5 9

Figure 4.4. (Case 2) Crushing injury to the skull of a 6-year-old. Diagrams of (a)
clinically observed fractures and (b) an overlay of the highest 20% of
principal tensile stresses produced ﬁ'om a FE model. White cranial
margins represent contact of applied forces (ground and tire) ..................... 60

Figure 4.5. (Case 3) Crushing injury to the skull of a 3-year-old. Diagrams of (a)
clinically observed fractures and (b) an overlay of the highest 20% of
principal tensile stresses produced from a FE model. White cranial
margins represent contact of applied forces (ground and tire) ..................... 60

Figure 4.6. (Case 4) Crushing injury to the skull of a 1.5 year old. Diagrams of (a)
clinically observed fractures and (b) an overlay of the highest 20% of
principal tensile stresses produced from a FE model. White cranial
margins represent contact of applied forces (ground and tire) ..................... 61

Figure 5.1. Simpliﬁed cranial model with overlays of modeled structures on true
structures. Coronal cut of CT used for thickness comparison. *Feature

viii

not modeled. TFeature simpliﬁed. #Suture width altered for consistency
with material pr0perty data obtained previously. Dotted line represents
location of frontal bone removal .................................................................. 71

Figure 5.2. Overlay of maximum principal tensile stresses from FE models on
representative experimental cranial fractures for (a) rigid interface
impact to a young specimen, (b) rigid interface impact to an old
specimen, (c) compliant interface impact to a young specimen, and (d)
compliant interface impact to an old specimen. Longer lines represent
higher stresses ............................................................................................... 73

Figure 5.3. Overlay of maximum principal tensile strains from FE models on
representative experimental cranial ﬁactures for (a) rigid interface
impact to a young specimen, (b) rigid interface impact to an old
specimen, (c) compliant interface impact to a young specimen, and (d)
compliant interface impact to an old specimen. Longer lines represent
higher strains ................................................................................................ 74

 

ix

RESEARCH PUBLICATIONS

PEER-REVIEWED MANUSCRIPTS

Baumer T, Nashelsky M, Hurst C, Passalacqua N, Fenton T, and Haut R.
Characteristics and Prediction of Cranial Crush Injuries in Children (in review)

Baumer T, Passalacqua N, Powell J, Newben'y W, Smith W, Fenton T, and Haut R.
Age-Dependent Fracture Characteristics of Rigid and Compliant Surface
Impacts on the Infant Skull - A Porcine Model. J. Forensic Sciences. 2009 (in
press)

Baumer T, Powell B, Fenton T, and Haut R. Age Dependent Mechanical Properties of
the Infant Porcine Skull and a Correlation to the Human. J Biomechanical Eng.
2009 (in press)

Meyer E, Baumer T, Slade J, Smith W, Haut R. Tibiofemoral contact pressures and
osteochondral microtrauma during ACL mpture due to excessive compressive
loading and internal torque of the human knee. Am J Sports Med. 2008; doi:
10.1 177/0363546508318046

PEER-REVIEWED ABSTRACTS

Baumer TG, Powell BJ, Fenton TW, Haut RC. Age Dependent Mechanical Properties of
the Infant Porcine Skull and a Correlation to the Human. Am. Society of
Mechanical Engineering, Lake Tahoe, CA. 2009.

Fenton TW, Passalacqua NV, Baumer TG, Powell BJ, Haut RC. A Forensic Pathology
Tool to Predict Pediatric Skull Fracture Patterns - Part 1: Investigations on Infant
Cranial Bone Fracture Initiation and Interface Dependent Fracture Patterns. Am.
Academy of Forensic Sciences (AAFS), Denver, Colorado, 2009. [Winner of the
Ellis R. Kerley Award]

Nashelsky MB, Fenton TW, Passalacqua NV, Hurst CV, Baumer TG, Haut RC. Cranial
Fracture Patterns in Pediatric “Crushing” Injuries and Preliminary Biomechanical
Modeling Using a Simple Finite Element Model. AAFS, Denver CO, 2009.

Meyer EG, Baumer TG, Haut RC. Knee joint relative motion during ACL rupture by
internal tibial torsion of tibiofemoral compression. North Am. Congress on
Biomechanics (NACOB), Ann Arbor, Michigan, 2008.

Meyer EG, Baumer TG, Haut RC. Tibiofemoral contact pressures and osteochondral
microtrauma from ACL mpture via hyperextension and joint compression.
NACOB, Ann Arbor, MI, 2008.

Meyer EG, Baumer TG, Haut RC. Tibiofemoral contact pressures and osteochondral
microtrauma during ACL rupture due to excessive compressive loading and
internal tibial torsion. NACOB, Ann Arbor, MI, 2008.

CHAPTER ONE

CHARACTERISTICS OF THE DEVELOPING SKULL - AN
INTRODUCTION

Skeletal ﬁ’actures are diagnosed in nearly 1 of 3 children investigated for
physical abuse with eighty percent of all fractures from abuse occurring in children less
than 18 months of age (Belfer et al, 2001). Head injuries account for 80% of fatal child
abuse (Case et al, 2001), while short falls rarely, if ever, cause serious injury or death
(Reiber, 1993). However, non-fatal injury from short falls (less than 3 feet) can still
result in skull fracture especially for infants less than 12 months of age (Gruskin and
Schutzman, 1999). Distinguishing between abuse and accidental trauma can be very
difﬁcult since both types of trauma may produce similar types of injuries (Billrnire and
Myers, 1985). Speciﬁcally, linear, complex, and depressed ﬁactures can be seen in both
types of cases (Reece and Sege, 2000; Wheeler and Shope, 1997). Some attempts have
been made to create a model of the developing skull, but data are limited. One modeling
effort involves scaling of the adult skull, which has met with some success in predicting
impact response of the pediatric skull (Prange et al, 2004). However, the head of an
infant is smaller and geometrically unlike that of an adult (Schneider et al, 1986) and the
validity of predicting skull fracture patterns in infants from adult data has not been
investigated.

Using adult data to predict skull fracture patterns in the pediatric skull may also
be problematic due to the different structural (ease of deformation and decreased
threshold to fracture) and mechanical properties of the infant skull (Thibault and

Margulies, 1998; Prange et al, 2000; Prange and Margulies, 2002). The developing skull

 

consists of several bones joined by soft tissue (sutures) that allow growth of the cranium
during childhood (Figure 1.1). These bones are developed during fetal life and grow
rapidly over the ﬁrst two years after birth, with decreasing growth rates into adulthood
(Mertz, 1984). The anterior fontanelle closes between 9 and 26 weeks after birth, but is
not sealed until around 18 months (Knight, 1991). In adulthood, the sutures of the skull
ossify creating one solid structure, whereas the sutures of infants result in more isolated

regions.

  
   
  
  
 
 

Frontal Bones Metopic Suture

Anterior Fontanelle
oronal Suture

Sagittal Suture

Parietal Bones
Posterior Fontanell
Occipital Bone Lambdoid Suture

Figure 1.1. Diagram of a newborn skull.

While several studies have documented important information regarding the
response of the infant skull and brain to blunt force impact, these data are limited to one
or two age groups. And because of ethical considerations, experimentation on human
pediatric specimens is very limited. However, one animal model of the human infant has

begun to emerge regularly in the current literature. A study by Dickerson and Dobbing

(1967) shows a similarity in development of the central nervous system when correlating
months of the human to weeks of the pig. Other studies have used porcine models to
predict fracture loads for the infant human femur (Pierce et al., 2000) and strain in the
braincase and sutures (Herring and Teng, 2000). An important ﬁnding in studies on the
porcine infant cranial bone and suture is “that the elastic modulus, rupture modulus, and
energy absorbed to failure of infant (2-3 day old) cranial (parietal) bone are similar to
that of the human infant in three-point bending” (Margulies and Thibault, 2000).

While Margulies and Thibault (2000) have attempted to correlate the mechanical
behavior of human infant cranial bone to porcine infant cranial bone, the ages and
number of specimens tested were limited. The data shows a linear increase in
mechanical properties with age, but includes prenatal and postnatal aged specimens.
Since questions of abuse in the pediatric patient typically involve subjects up to 18
months of age, which correlates to a time of rapid skull growth, the biomechanical
properties of developing skull bones and sutures must be well documented over this
entire period of developmental time if an accurate age-dependent model is to be created.
Yet, the age-dependent properties of the developing skull are not the only conditions
affecting fracture patterns. The risk of injury is also dependent on the contacting surface
and fall height (Bertocci et al., 2003; Chalmers et al., 1996). Other variables, such as the
area struck, thickness of the skull, thickness of scalp and hair, and impact direction can
also affect the fracture pattern (Knight, 1991; Cooperrnan and Merten, 2001). Head
injuries are often the result of dynamic forces in which a moving object strikes the
cranium at a considerable speed or the cranium makes contact with a static object that

causes the skull to be rapidly decelerated (Duhaime et al., 1995). However, cases of

 

crushing injuries, where a static load is applied relatively slowly to the cranium, are also
present in the literature. Unfortunately, due to the paucity of skull fracture tolerance data
for infants and young children, pediatric trauma involving single-event head injuries
with related cranial fractures represent one of the greatest challenges to forensic
pathologists and anthropologists. Since it is often the job of the forensic professional to
try to recognize characteristic patterns in these cases, a clear understanding of fracture
propagation, and the effects of the aforementioned variables, becomes vital.

Studies involving impact loading of the cranium have been presented, but there
are also several case reports of basilar skull damage in the literature involving both
crushing and impulse based loading scenarios mainly focusing on adult crania. A study
involving 15 cases of crushing injuries to adult skulls aged 14-63 years and
experimentally induced ﬁactures on 11 cadaver crania revealed similarities in basilar
fracture patterns (Russell and Schiller 1949). However, a study by Harvey and Jones
(1979) claimed that fractures through both petrous bones was not an indicator of impact
site. Another study by Yasue (1981) then suggested that the overall path of fracture
through the basicranium could, in fact, be related to the region of the skull impacted.
While it is assumed that tinctures will generally follow the path of least resistance, this
path may be different between the child and adult cranium. While all of these
aforementioned studies focused on adult crania, some recent case reports on quasi-static
loading to the crania of children have also presented observations of basilar ﬁactures
(Takeshi, 2005; Duhaime et al., 1995). The trend of quasi-static loading producing
basilar fractures in both adults and children may suggest that prediction of these

fractures is possible.

Theoretical techniques are widely used for prediction of material failure in
structures. The Finite Element (FE) method is a powerful analytical modeling tool that
can be used to determine internal stresses and strains within a complex solid. Several
studies have shown that maximum principal stress (or strain) directions from a modeled
structure correlate well with areas of experimentally or clinically observed bone
fractures (Graham 2000; Silva 1998; Doorly 2006). While modeling of ﬁacture
propagation generally involves high computational costs, one theory suggests that there
is a close, although not exact, relationship between pre-failure stress trajectories and the
ultimate path of propagation. In brief, Frank and Lawn (1967) proposed that crack
growth occurs in such a way as to maximize the energy dissipation per unit length of
growth, and that the greatest amount of dissipation occurs when the crack plane is
perpendicular to the direction of greatest principal tensile stress. In other words, the
directions of principal tensile stress developed on a solid may be a reasonable predictor
of propagation direction.

Skull fracture interpretation is a necessary component of forensic analysis and
contributes to the determination of the cause and manner of injury or death. The research
presented in this thesis makes use of an in situ porcine model and ﬁnite element analysis
to examine the responses of the developing skull to blunt force impulse and quasi-static
crushing trauma. Chapter 2 documents age effects on the mechanical properties of
parietal bone and coronal suture in porcine infants and correlates the bending properties
of the bone to existing human infant data. Chapter 3 documents skull ﬁ-acture on infant
porcine specimens with respect to age and interface in an effort to identify fracture

characteristics for a developing skull under known conditions. This study hypothesized

-A..

that impact energy required to initiate a skull fracture would depend on specimen age,
and the extent of fracture damage for a given energy impact would depend on impact
interface. Chapter 4 uses a simpliﬁed cranial model and the ﬁnite element method to
develop representative patterns of principal stress and strain directions developed during
quasi-static crushing injuries to the developing skull. The theoretical results were
compared to four clinical cases of known childhood fatalities from crushing head
injuries inﬂicted by vehicle tires. Chapter 5 uses a simpliﬁed porcine cranial model and
the ﬁnite element method to predict experimentally documented fracture patterns
developed dming impact loading to the parietal bone. The material properties of the
cranial model were taken from the previously obtained results from Chapter 2. The
experimentally inﬂicted fractures to the cranium from Chapter 3 were used for
comparison with the theoretically developed principal stress and strain directions.

The research presented in this thesis provides useful data in regards to the
material properties of the developing porcine skull, and has suggested a correlation with
the developing human skull. This research also documents fractures generated on the
developing skull due to blunt force trauma with respect to age and interface. The
experimentally and clincally observed ﬁ'actures were predicted using a simpliﬁed
theoretical model of impact and quasi-static loading of the cranium. This information
may prove relevant in determining perimortem events in forensic cases of pediatric
cranial injuries with unknown or questionable circumstances. These data may have

ultimate utility in the study of potential child abuse cases.

REFERENCES

Belfer R, Klein B, and Orr L., 2001, “Use of the Skeletal Survey in the Evaluation of
Child Maltreatment,” American Journal of Emergency Medicine, 19, pp. 122—4.

Bertocci G. and Pierce M., 2006, “Applications of Biomechanics Aiding in the
Diagnosis of Child Abuse,” Clinical Pediatric Emergency Medicine, 7, pp. 194-9.

Billmire M. and Myers P., 1985, “Serious head injury in infants: accident or abuse?,”
The American Academy of Pediatrics, 75, pp. 340-2.

Case M, Graham M, Handy T, Jentzen J, Monteleone J ., 2001, “Position Paper on Fatal
Abusive Head Injuries in Infants and Young Children,” The American Journal of
Forensic Medicine and Pathology, 22, pp. 112-22.

Chalmers D, Marshall S, Langley J, Evans M, Brunton C, Kelly A, et al., 1996, “Height
and surfacing as risk factors for injury in falls from playground equipment: a case-
control study,” Injury Prevention, 2, pp. 98-104

Cooperman D. and Merten D., 2001, “Skeletal Manifestation of Child Abuse,” In: Child
Abuse: Medical Diagnosis and Management. Philadelphia: Lippincott, Williams, and
Wilkins, pp. 123-56.

Dickerson J. and Dobbing J ., 1967, “Prenatal and postnatal growth and development of
the central nervous system of the pig,” Proceedings of the Royal Society of London,
166(1005), pp. 384-95.

Doorly M, Gilchrist M., 2006, “The use of accident reconstruction for the analysis of
traumatic brain injury due to head impacts arising from falls,” Computer Methods in
Biomechanics and Biomedical Engineering, 9(6), pp. 371-7.

Duhaime A-C, Eppley M, Margulies S, Heher K, Bartlett S., 1995, “Crush Injuries to the
Head in Children,” Neurosurgery, 37(3), pp. 401-7.

Frank F, Lawn B., 1967, “On the Theory of Hertzian Fracture,” Proceedings of the
Royal Society of London, 299(1458), pp. 291-306.

Graham R, Oberlander E, Stewart J, Griffiths D., 2000, “Validation and use of a ﬁnite
element model of C-2 for determination of stress and fracture patterns of anterior
odontoid loads,” Journal of Neurosurgery (Spine 1), 93, pp. 117-25.

Gruskin K. and Schutzrnan S., 1999, “Head Trauma in Children Younger Than 2 Years,”
Archives of Pediatric Adolescent Medicine, 153, pp.15-21.

Harvey P, Jones A., 1980, “‘Typical’ Basal Skull Fracture of Both Petrous Bones: An
Unreliable Indicator of Head Impact Site,” Journal of Forensic Sciences, 25(2), pp. 280-
6.

Herring S. and Teng S., 2000, “Strain in the Braincase and Its Sutures During Function,”
American Journal of Physical Anthropology, 112, pp. 575-593.

Knight 3., 1991, “Fatal child abuse,” In: Forensic Pathology. London: Edward Arnold,
pp. 457-73.

Mertz H., 1984, “A Procedure for Normalizing Impact Response Data,” SAE #840884,
29th Stapp Car Crash Conference, Chicago, IL.

Pierce M, Valdevit A, Anderson L, Inoue N, and Hauser D., 2000, “Biomechanical
Evaluation of Dual-Energy X-Ray Absorptiometry for Predicting Fracture Loads of the
Infant Femur for Injury Investigation: An In Vitro Porcine Model,” Journal of
Orthopaedic Trauma, 14(8), pp. 571-6.

Prange M, Luck J, Dibb A, Van Ee C, Nightingale R, Myers B., 2004, “Mechanical
Properties and Anthropometry of the Human Infant Head,” Stapp Car Crash Journal, 48,
pp. 279-99. -

Prange M, Meaney D, Margulies S., 2000, “Deﬁning brain mechanical properties:
Effects of region, direction, and species,” Stapp Car Crash Journal, 44, pp. 205-213.
Prange M. and Margulies S., 2002, “Regional, Directional, and Age-Dependant
Properties of the Brain Undergoing Large Deformations,” Journal of Biomechanical
Engineering, 124(2), pp. 244-52.

Reece R and Sege R., 2000, “Childhood Head Injuries: Accidental or Inﬂicted,”
Archives of Pediatric Adolescent Medicine, 154, pp. 11-15.

Reiber G., 1993, “Fatal falls in childhood: How Far Must Children Fall to Sustain Fatal
Head Injury?,” American Journal of Forensic Medicine and Pathology, 14, pp. 201-7.

Russell W, Schiller F ., 1949, “Crushing Injuries to the Skull: Clinical and Experimental
Observations,” Journal of Neurology, Neurosurgery, and Psychiatry, 12, pp. 52-60.

Schneider L, Lehman R, Pﬂug M, Owings C., 1986, “Size and Shape of the Head and
Neck From Birth to Four Years,” Washington, DC. The Consumer Product Safety
Commission, Report No.: UMTRI—86-2.

Silva M, Keaveny T, Hayes W., 1998, “Computed Tomography-Based Finite Element
Analysis Predicts Failure Loads and Fracture Patterns for Vertebral Sections,” Journal of
Orthopaedic Research, 16, pp. 300-8.

Takeshi M, Okuchi K, Nishiguchi T, Seki T, Watanahe T, Ito S, et al., 2006, “Clinical
Analysis of Seven Patients of Crushing Head Injury,” The Journal of Trauma-Injury,
Infection, and Critical Care, 60(6), pp. 1245-9.

Thibault K and Margulies S., 1998, “Age-Dependant Material Properties of the Porcine
Cerebrum: Effect on Pediatric Head Injury Criteria,” Journal of Biomechanics, 31(12),
pp. 11 19-26.

Wheeler D. and Shope T., 1997, “Depressed skull fracture in a 7-month old who fell
from bed,” Pediatrics, 100, pp. 1033-4.

Yasue M., 1981, “Clinical and experimental investigation of mediobasal skull fracture,”
Neurol Med Chir (Tokyo), 21, pp. 1041-9.

CHAPTER TWO

AGE DEPENDENT MECHANICAL PROPERTIES OF THE INFANT
PORCINE PARIETAL BONE AND A CORRELATION TO THE
HUMAN

ABSTRACT

An infant less than 18 months of age with a skull fracture has a 1 in 3 chance of abuse.
Injury biomechanics are often used in the investigation of these cases. In addition to
case-based investigations, computer modeling, and test dummies, animal model studies
can aid in these investigations. This study documents age effects on the mechanical
properties of parietal bone and coronal suture in porcine infants and correlates the
bending properties of the bone to existing human infant data. Three beam specimens
were cut from porcine specimens aged 3, 7, 10, 14, 18, and 21 days: one across the
coronal suture and two from the parietal bone, one parallel to and one perpendicular to
the coronal suture. An actuator-mounted probe applied 4-point bending in displacement
control at 25nmr/sec until failure. Bending stiffness of bone specimens increased with
age; bone-suture—bone specimens showed no change up to 14 days, but increased from
14 to 18 days. All three specimen types showed decreases in ultimate stress with age.
Ultimate strain for the bone-suture-bone specimens was signiﬁcantly higher than that for
the bone specimens up to 14 days, with no differences thereafter. There was no change
in the bending modulus with age for any specimen type. Bone-suture-bone bending
modulus was lower than that of the bone specimens up to 14 days, with no differences
thereafter. There was no change in strain energy to failure with age for the bone
specimens; bone-suturebone specimens showed no change up to 14 days, but decreased

from 14 to 18 days. There was an increase in specimen porOsity with age. Correlation

10

 

analysis revealed a weak (-0.39), but signiﬁcant, negative correlation between ultimate
stress and porosity. While the mechanical properties of parietal bone and coronal suture
did not change signiﬁcantly with age, bone specimens showed an increase in bending
stiffness with age. Bone-suture—bone specimens showed an increase in bending stifﬁress
only between 14 and 18 days of age. Correlation analyses using existing and new data to
compute the bending rigidity of infant parietal bone specimens suggested that days of
pig age may correlate with months of human age during the most common time frame of
childhood abuse cases. Such a correlation may support the developing porcine model as
an important biomechanical tool for the study of pediatric skull ﬁacture in efforts to

investigate cases of childhood abuse.

11

INTRODUCTION

Skeletal fractures are diagnosed in nearly 1 of 3 children investigated for
physical abuse (Belfer et al., 2001). Often it is the job of the forensic anthropologist to
try to recognize characteristic patterns in these cases, but in reality it is difﬁcult to
diagnose abuse simply based on a fracture. In fact, age together with the type and site of
tincture are important features to consider when assessing cases of potential abuse
(Kemp et al., 2008). Eighty percent of all ﬁactures from abuse are seen in children less
than 18 months of age. For these children the probability of abuse given a skull fracture,
for example, is 0.30 (range 0.19 to 0.46). The most common site of skull fracture in
abuse and non-abuse cases is the parietal bone (Meservy et al., 1987), with the most
common fracture type being linear (Reece and Sege, 2000).

Injury biomechanics are often used in the investigation of cases suspected to
involve child abuse (Bertocci and Pierce, 2006). In addition to case-based investigations,
computer modeling, and the use of test dummies, animal model studies can aid in these
investigations. While the relationship between animal studies and the human pediatric
patient is yet unclear in most circumstances, one animal model seems to have regularly
emerged in the current literature. For example, a study has been conducted on prediction
of fracture loads for the infant femur using porcine models aged 3-12 months (Pierce et
al., 2000). In another study on strain in the braincase and its sutures, juvenile pigs have
again been used as a model of the human infant (Herring and Teng, 2007). Dickerson
and Dobbing (1966) have previously compared the development of the porcine central
nervous system to the human and found that months of life in the human are roughly

comparable to weeks in a pig. An important ﬁnding in studies on the porcine infant

12

cranial bone and suture is “that the elastic modulus, rupture modulus, and energy
absorbed to failure of infant (2-3 day old) cranial (parietal) bone are similar to that of the
human infant in three-point bending” (Margulies and Thibault, 2000).

While Margulies and Thibault (2000) have attempted to correlate the mechanical
behavior of human infant cranial bone to porcine infant cranial bone, the ages and
number of specimens tested were limited. The data shows a linear increase in
mechanical prOperties with age, but the data includes prenatal and postnatal aged
specimens. Other studies have shown that growth of the human skull involves large
increases in size over the ﬁrst two years of life, with decreasing growth rates into
adulthood (Mertz, 1984). Since questions of abuse in the pediatric patient typically
involve subjects up to 18 months of age, an objective of the current study was to
document the biomechanical properties of developing porcine skulls and sutures over a
longer period of developmental time than previous studies.

Additionally, Margulies and colleagues have documented the biomechanical
properties of the infant porcine parietal bone using beam specimens oriented
perpendicular to and across the coronal suture, but case-based investigations show linear
skull fractures in infants typically initiate at an ossiﬁcation center and run parallel to and
perpendicular to spiculae in the parietal bone (Holck, 2005). It has also been shown that
the mechanical properties of prenatal human infant parietal bone is non-isotropic
(McPherson and Kriewall, 1980). Therefore in the current study, beam specimens were
taken from the developing porcine parietal bone both parallel and perpendicular to the

coronal suture, as well as across it. The data ﬁom the current study will provide

13

information on the age factor for developing parietal bone and coronal suture in the
porcine infant model, as it may relate to the human pediatric patient.
MATERIALS AND METHODS

Porcine specimens aged 3 days (n=5), 7 days (n=8), 10 days (n=7), 14 days
(n=5), 18 days (n=5), and 21 days (n=4) were obtained from a local supplier and stored
at —20°C until testing. All animals died of natural causes and the specimens were frozen
within 12 hrs of death. None of the specimens had outward signs of head injury, which
was later conﬁrmed during specimen preparation.

Each specimen was thawed at room temperature for 24 hours prior to testing. The
scalp was removed, but the periosteum was leﬁ intact to help prevent drying of the bone
and sutures. The top palette of the skull was removed using a surgical saw (Autopsy
Saw, Stryker, Kalamazoo, MI). Cuts were made at the larnbdoid suture, along the lateral
aspect of the parietal bone, and diagonally across the home] bone from the lateral aspect
of the coronal suture to an anterior location along the sagittal suture.

Three beam specimens were cut from each skull: one across the coronal suture
and two from the parietal bone, one parallel to and one perpendicular to the coronal
suture (Figure 2.1). The specimens were cut with a bone saw (Isomet 1000, Buehler,
Lake Bluff, IL), while bathed in phosphate buffered saline (PBS). The specimen width
varied from 4.5 to 5.5 mm wide. Due to the non-uniform thickness of the beam
specimens, three thickness measurements were taken from each specimen at equally
spaced intervals along the beam. These measurements were used to calculate an average

thickness of each specimen. The length of the beam specimens was ﬁxed at 15 mm to

14

yield a length to thickness ratio of approximately 8:1. The cut specimens were stored in

room-temperature PBS for less than 1 hour before testing.

Sagittal
Suture

  
  
 
  
  

Bone-Suture-
Bone

Coronal
Suture

 

Parallel Perpendicular
Occipitall

Figure 2.1. TOp view of cranium with locations of the beam specimens.

The beam specimens were loaded to failure in 4-point bending. Each specimen
was mounted between two aluminum sleeves using air-hardening cement (Technovit,
J orgensen laboratories Inc., Loveland, CO). The ﬁxed testing length of 15 mm was
maintained with a thin aluminum bar fastened between the sleeves. This bar also served
to prevent damage to the specimen during preparation and mounting in the test ﬁxture.

The potted specimens were mounted in a custom-designed 4-point bending
ﬁxture (Figure 2.2) attached to a servo-hydraulic testing machine (Model 1331, Instron
Corp., Canton, OH). One end of the ﬁxture was mounted on linear bearings to allow
displacement along the specimen axis. An actuator-mounted probe was located on each
of the two sleeves to apply 4-point bending. A 111.2-N load-cell (Model #JP25, Data
Instruments Inc., Acton, MA) was mounted between the probe and actuator. The
experiments were run with the machine in displacement control at 25mm/sec to generate

the bending moment on the beam specimens.

15

Linear Actuator

 

 

 

 

 

 

 

 

 

 

Load Cell
Probe
O _l:::|_ O
[: Beam ]
Specimen
i c— Translation —> i

 

 

 

 

 

| J

Figure 2.2. Diagram of testing ﬁxture.

The force and actuator displacement were recorded to failure during each

experiment. From these values, initial specimen stifﬁress, ultimate stress (cult): ultimate

strain (a uh), bending modulus (E), and strain energy to failure (U) were determined

using the following information. Initial stifﬁress was calculated from the initial linear
portion of the force-actuator displacement plot. Ultimate stress at the endocranial surface
was determined ﬁom simple beam theory using
M*
_ 3’
Cult _ 1
f
where M. was the bending moment at failure, y was the distance from the inner surface

(2.1)

 

of the specimen to the centroidal axis, and If was the second moment of the rectangular

area measured at the ﬁacture location. Ultimate strain at the fracture location was
calculated from

s = —- (2.2)

where y was the distance from the inner surface of the specimen to the centroidal axis, or
was the angular rotation of the beam under the probes at failure, and L was the length of

the specimen. The elastic modulus was also determined ﬁ‘orn simple beam theory using

__rr L

__ (23)
a 21
av

E

 

g
where War was the moment-angular rotation ratio during the initial linear region of the

force-displacement trace, and Iavg was the second moment of area determined using the

average thickness over the length of the beam specimen. The failure strain energy was

then computed with the following integral equation

2
LENF)
U:
I2E1
0 avg

where ‘x’ was the distance along the beam from the stationary aluminum cylinder.

dx (24)

 

The gross morphological features of the beams were documented by staining the
length-wise cross-section of the failed specimens with India ink (Figure 2.3). Each test
specimen was photographed with a digital camera mounted to a dissecting microscope at
50x magniﬁcation (Wild M5A, Wild Heerbrugg, Switzerland). A representative section
of bone was cropped ﬁ'om the exocranium to the endocranium that included an equal
distance along the length of the beam. This image was always taken in an area between
characteristic undulations in the specimen thickness, as this was the area of failure
during the experiments. The image was imported to an image analysis program (Sigma
Scan, J andel Scientiﬁc, San Rafael, CA) and converted to grayscale. The porosity was

determined as the area of the stain (voids) per total area.

17

Compact Layer

Porous Layer

 

Figure 2.3. Representative stained cross-section of a beam from a 14 day specimen
showing separation of compact and porous layers. Voids in the bone matrix were colored
black. The box represents the analyzed area of a test specimen.

Each material property data set was analyzed for age effects using linear
regression analyses. Comparisons between the three specimen types was performed
using a two-factor (age, type) AN OVA. A Pearson Product Moment correlation was
used to analyze the effect of porosity on ultimate stress. Statistically signiﬁcant effects
were reported for p<0.05.

RESULTS

There was no difference in average thickness measurements between the bone
and bone-suture-bone specimens for any age. Linear regression analyses also showed a
signiﬁcant increase in thickness with age (p<0.001 for each specimen type).

For each orientation of parietal bone specimen, there was a signiﬁcant increase
in bending stifﬁress with age (Figure 2.4a). The slopes were not statistically dependent
on orientation. Interestingly, there was no signiﬁcant change in bending stiffness in the
bone-suture-bone specimens up to fourteen days, but between fourteen and eighteen
days there was a signiﬁcant increase (Figure 2.4b). The stiffness of the suture was lower
than that of the bone up to fourteen days, but it became statistically greater than the

parallel bone specimens at eighteen days (suture 1197i246 N/m; parallel 864d:367 N/m).

18

2500 — - Perpendicrlar a Parallel
2000 4 n
1500 -
1000 -

Stiffness (N/m)

 

 

UT
:3 53
r

 

O
(a)
O)
(D

12 15 18 21
Age (days)

A
m
v

2000 -

 

_s .3
N O)
C O
O C
r r

800 -
400 .

Stiffness (N/m)

 

 

3 7 10 14 18 21
(b) Age (Days)

Figure 2.4. Recorded values of stiffness for (a) both orientations of bone and (b) bone-
suture-bone specimens from 3 to 21 days. *Signiﬁcantly different means.

All bone specimens failed at a thinned area and all bone-suture-bone specimens
failed in the suture. Linear regression analyses showed signiﬁcant decreases in ultimate
stress for both orientations of bone (p<0.05) and for the bone-suturebone specimens
(p<0.01) with age (Figure 2.5). The slopes were not different.

There was no change with age in ultimate strain in the bone (0.0163:I:0.0053
mm/mm) and bone-suture-bone (0.0313i0.0072 mm/mm) specimens (Figure 2.6). The
ultimate strain for the bone-suture-bone specimens was signiﬁcantly higher than that for

the bone specimens up to fourteen days.

19

1000 - o Suture

E I I Perpendicular
E 300 ‘ a Parallel

g 600 -

a

g 400 -

g 200 -

:42

3 o

 

 

 

0 3 6 9 12 15 18 21

    

Age (days)
Figure 2.5. Linear regressions of ultimate stress for bone and bone-suture-bone
specimens against age.
0.06 - * * FL I Suture
’T‘ T ‘1" I Perpend.
.g 0-05 ' re "" |'_'l Parallel
5 0.04 -
03 ..
g 0.03
E 0.02 -
3 0.01 -
0 -

 

3 7 10 14 18 21
Age (Days)

Figure 2.6. Ultimate strain for bone and bone-suture-bone specimens. *Signiﬁcantly
different means.

There was no change with age in the bending modulus of the bone (6.01:1:1 .73
GPa) and bone-suture-bone (2.731084 GPa) specimens (Figure 2.7). The bending
modulus of the bone-suture-bone specimens was signiﬁcantly lower than that of the

bone specimens up to fourteen days.

20

I Suhte
,, I Perpend.
"H * * El Parallel

Bending Modulus (GPa)

 

 

 

 

3 7 1 0 14 1 8 21
Age (Days)
Figure 2.7. Bending modulus for bone and bone-suture-bone specimens versus age.

*Signiﬁcantly different means.

There was also no change with age in the strain energy to failure for the bone
specimens, being 0.03:l:0.03 J and 0.03:1:004 J for perpendicular and parallel
orientations, respectively (Figure 2.8). The bone-suture-bone specimens showed no
change up to fourteen days (02310.13 J). For ages up to fourteen days, the strain energy
to failure of the bone-sutrrre-bone specimens was signiﬁcantly higher than that of the
bone specimens. From fourteen to eighteen days, however, there was a signiﬁcant
decrease in the strain energy of the bone-suture-bone specimens. Therefore, the strain
energy to failure was no longer different for the bone (0.01:1:000 J) versus bone-suture-
bone specimens (0.01:l:0.00 J).

The morphological characteristics of the skull bone ﬁom the infant porcine
specimens changed during the aging process. It was found that two distinct layers, one
that was visually quite compact and another more porous layer, were present as early as

three days. The porous layer appeared as pockets beneath the compact layer which

21

developed through ten days, and by fourteen days there was a clear separation of zones

along the entire specimen (see Figure 2.3).

Strain Energy (J)

0.6 -
0.5 d
0.4 -
0.3 -
0.2 -
0.1 -

 

0..

I Sutue
* * I Perpend.
T T t * El Parallel
re A

 

3 7 1o 14 1a 21
Age(DaVS)

Figure 2.8. Stain energy of bone and bone-suture bone specimens versus age.
*Signiﬁcantly different means.

Analysis of both orientations of bone specimens revealed an increase in porosity

with age (p<0.001) (Figure 2.9). Correlation analysis of the combined data revealed a

weak (-0.391), but signiﬁcant (p<0.01), negative correlation between ultimate stess and

porosity. No other mechanical property had a signiﬁcant correlation with porosity.

ItY

Poros

20% -

10% -

0%

40% - I Perpendicular :- Parallel

30% - '

 

 

 

0 3 6 9 12 15 18 21
Age(Days)

Figure 2.9. Linear regression analyses of both orientations of infant porcine bone
specimens against age. Both orientations displayed signiﬁcant dependencies on age.

22

DISCUSSION

The current study was performed to document potential changes in the
mechanical properties of infant porcine parietal bone as a function of age. Linear
regression analyses on the bending modulus of porcine specimens showed no signiﬁcant
change with age. A few previous studies on human specimens have shown a linear
increase in the bending modulus of prenatal human infant parietal bone and it has been
suggested that this tend may continue up to six months (Margulies and Thibault, 2000),
or even six years (McPherson and Kriewall, 1980) of age. These studies, however,
assumed a linear increase from the prenatal data to the older specimens and did not
collect data between the endpoints. A study by Coats and Margulies (2006) investigates
the mechanical properties of both prenatal and postratal infants from twenty-six weeks
of gestation to thirteen months of age. The authors claimed that there was a signiﬁcant
increase in bending modulus with age, but a closer inspection of their data indicated that
the increase in these properties was signiﬁcantly inﬂuenced by the prenatal data. When
linear regression analysis was performed on the postnatal data alone there was no
statistical change in bending modulus with age. These data are in agreement with the
current study using infant porcine skull bone.

The bending modulus data tom the human parietal bone of Coats and Margulies
(2006) have also been compared to the bending modulus of the porcine parietal bone. It
was clear that the porcine parietal bone bending modulus was signiﬁcantly higher than
that of the human parietal bone. However, this is a material rather than a stuctural
parameter of comparison. Since the current thought was to use the porcine model to

study issues of potential childhood abuse, 3 stuctural comparison may be more

23

appropriate. In this effort we were able to obtain unpublished human infant parietal bone
thickness data for ages two to twenty-one months, n=314, ﬁ‘om a recent study by Li et al.
(2007). While the human skull was signiﬁcantly thicker than the porcine skull over the
ages studied here, a linear regression of parietal bone thickness versus age indicated
growth rates of parietal bone thickness for humans to be 0.0487 mm/month and that of
pigs to be 0.0492 mm/day. Using bending modulus data from Coats and Margulies
(2006) and thickness values obtained during the Li et al. (2007) study, the bending
rigidity of human infant parietal bone versus age in months was compared to that of the

porcine specimens versus age in days (Figure 2.10). The slopes of parietal bone bending

rigidity versus age were 0.505 GPa*m4/month for humans and 0.596 GPa*m4/day for

the porcine specimens. The difference between human and porcine data may be due to
differences in the stain rates of testing. The human specimens were tested at a high

velocity (approximately 2 m/sec), while the porcine specimens were tested at a lower

25 - 0 Human o Porcine

Bending Rigidity

 

 

 

Figure 2.10. Bending rigidity of human and porcine specimens versus age (Human —
months; porcine - days). The human data was based on data from Coats and Margulies
(2006) and unpublished data from Li et a1 (2007).

24

velocity (0.025 m/sec). While a previous study by Wood (1971) shows that the bending
modulus of adult human cranial specimens between twenty-ﬁve and ninety-ﬁve years of
age increases with increases in stain rate, the previous study by Coats and Margulies
(2006) showed no stain rate dependency in bending modulus with age for specimens
less than one month of age. Funn‘e studies may be needed to more ﬁrlly examine rate
dependencies in both the human and porcine infant cranial bone during early
developmental ages.

The current study was also performed to document potential changes in the
mechanical properties of the infant porcine coronal suture as a function of age.
Margulies and Thibault (2000) show a signiﬁcant difference between the bone-suture-
bone and bone specimen mechanical properties for two to three day old porcine
specimens. The data tom the current study was in agreement with this ﬁnding up to
fourteen days. Speciﬁcally, in the current study, the bone-suture-bone specimens were
less stiff, absorbed more energy to failure, and withstood larger bending deformation
before failure than the bone specimens. The differences between bone-suture-bone
specimens and bone specimens changed signiﬁcantly in the current study between
fourteen and eighteen days. During this period of development, the bone-suture-bone
specimens underwent a signiﬁcant increase in stiffness, a signiﬁcant decrease in the
energy absorbed to failure, and a decrease in stain at failure. These changes resulted in
mechanical properties of the bone-suture-bone specimens that were more similar to those
of the bone specimens at eighteen and twenty-one days. The explanation for this change
between fourteen and eighteen days is currently unclear tom visual analysis of the

specimen cross-sections. This could also be related to changes in the biochemical make-

25

up of the suture itself during this period. Some gross changes in the morphology of the
porcine bone did appear to occur over the time period of the current study. One
important feature was the development of two distinct layers of bone. Examination of
each cross-section revealed isolated pockets of porous bone beneath an outer, more
compact layer of bone in specimens as young as three days. By fourteen days, a clear
separation between the compact and porous layers could be seen over the entire cross-
section of these specimens. Previous studies on human (McElhaney et al., 1970; Behrens
et al., 1974) and bovine (Martin, 1984) samples have shown that small increases in
porosity of bone specimens can result in large decreases in their ultimate stess. The
Coats and Margulies (2006) study also documents an increase in ultimate stess with
specimen age, but our analysis of postnatal data alone out to thirteen months again
revealed no statistical change with age. These observations warrant additional studies on
the morphological development of human and porcine infant parietal bone.

The current study also showed no signiﬁcant difference in orientation on the
bending properties of parietal beam specimens. This ﬁnding was in contast to
previously published data for human specimens up to 40 weeks of gestation (McPherson
and Kriewall, 1980). Interestingly, the aforementioned study described a distinct ﬁber
orientation in both pre—term and term human infant parietal bones, but visual observation
of the developing infant porcine skull did not show any distinct spiculae that might be
related to a ﬁber orientation effect. While this ﬁnding seemed to reveal an inconsistency
between infant human and porcine skulls, we were not able to ﬁnd deﬁnitive information

on when this visible ﬁber orientation effect disappears in humans.

26

In summary, the current investigation documented alterations in the mechanical
properties of beam specimens tom the porcine infant skull with age. While it remains
difﬁcult to correlate these changes with those that may occur in the human pediatric
patient, due to limited human pediatric data, the current study has attempted to make a
correlation by combining existing and new, unpublished data. A good correlation
appeared to exist for days of pig age and months of human age in terms of ﬂexural
rigidity of the beam specimens. Such a correlation may lend support to the notion that
studies on the developing porcine model may provide an important biomechanical tool

for the study of pediatric skull tacture in efforts to investigate cases of childhood abuse.

27

REFERENCES

Behrens J, Walker P, and Shoji H., 1974, “Variations in Stength and Stucture of
Cancellous Bone at the Knee,” Journal of Biomechanics, 7, pp. 201-7.

Belfer R, Klein B, and Orr L., 2001 , “Use of the Skeletal Survey in the Evaluation of
Child Malteatnent,” American Journal of Emergency Medicine, 19, p. 122—4.

Bertocci G. and Pierce M., 2006, “Applications of Biomechanics Aiding in the
Diagnosis of Child Abuse,” Clinical Pediatric Emergency Medicine, 7, pp. 194-9.

Coats B. and Margulies S., 2006, “Material Properties of Human Infant Skull and Suture
at High Rates,” Journal of Neurotrauma, 23(8), pp. 1222-32.

Dickerson J. and Dobbing J ., 1966, “Some Peculiarities of Cerebellar Growth in Pigs,”
Procwdings of the Royal Society of Medicine, 59, pp. 1088.

Herring S. and Teng S., 2000, “Stain in the Braincase and Its Sutures During Function,”
American Journal of Physical Anthropology, 112, pp. 575-93.

Holck P., 2005, “What Can a Baby’s Skull Withstand? Testing the Skull’s Resistance on
an Anatomical Preparation,” Forensic Science International, 151, pp. 187-91.

Kemp A, Dunstan F, Harrison S, Morris S, Mann M, Rolfe K, Datta S, Thomas D, Sibert
J, and Maguire S., 2008, “Patterns of Skeletal Fractures in Child Abuse: Systematic
Review,” B.M.J., 337, pp. 1-8.

Li H, Ruan J, Xie Z, Wang H, and Liu W., 2007, “Investigation of the Critical
Geometric Characteristics of Living Human Skulls Utilising Medical Image Analysis
Techniques,” Int. J. Vehicle Safety, 2(4), pp. 345-61.

Margulies S. and Thibault K., 2000, “Infant Skull and Suture Properties: Measurements
and Implications for Mechanisms of Pediatric Brain Injury,” Journal of Biomechanical
Eng'neering, 122, pp. 364-71.

Martin R., 1984, “Porosity and Speciﬁc Surface of Bone,” Critical Reviews in
Biomedical Engineering, 10, pp. 179-222.

McElhaney J, Fogle J, Melvin J, Haynes R, Roberts V, and Alern N., 1970, “Mechanical
Properties of Cranial Bone,” Journal of Biomechanics, 3, pp. 495-511.

McPherson G. and Kriewall T., 1980, “The Elastic Modulus of Fetal Cranial Bone: A

First Step Towards an Understanding of the Biomechanics of Fetal Head Molding,”
Journal of Biomechanics, 13, pp. 9-16.

28

Mertz H., 1984, “A Procedure for Normalizing Impact Response Data,” SAE #840884,
29th Stapp Car Crash Conference, Chicago, IL.

Meservy C, Towbin R, McLaurin R, Myers P, and Ball W., 1987, “Radiographic
Characteristics of Skull Fractures Resulting From Child Abuse,” A.J.R., 149, pp. 173-5.

Pierce M, Valdevit A, Anderson L, Inoue N, and Hauser D., 2000, “Biomechanical
Evaluation of Dual-Energy X-Ray Absorptiomety for Predicting Fracture Loads of the
Infant Femur for Injury Investigation: An In Vito Porcine Model,” Journal of
Orthopaedic Trauma, 14(8), pp. 571-6.

Reece R. and Sege R., 2000, “Childhood Head Injuries: Accidental or Inﬂicted,”
Archives of Pediatric Adolescent Medicine, 154, pp. 11-15.

 

Wood J ., 1971, “Dynamic Response of Human Cranial Bone,” Journal of Biomechanics,
4, pp. 1-12.

29

CHAPTER THREE

AGE-DEPENDENT FRACTURE CHARACTERISTICS OF RIGID AND
COMPLIANT SURFACE IMPACTS ON THE INFANT SKULL — A
PORCINE MODEL

ABSTRACT

Head injuries account for 80% of fatal child abuse in younger children. However, non-
fatal injury ﬁ'om short falls (less than 3 feet) can still result in skull fracture especially
for infants less than 12 months of age. In many cases the ability to determine child abuse
from accidental injury may be difﬁcult. This study documents skull fracture on infant
porcine specimens with respect to age and interface in an effort to identify ﬁacture
characteristics for a developing skull under known conditions. A single impact causing
ﬁacture was conducted on the skull of porcine specimens aged 2 to 28 days (n=76).
Impacts were conducted using a gravity-accelerated mass to deliver a contolled energy
impact to the parietal bone. Paired rigid and compliant impacts at the same energy were
conducted at each specimen age. Impact force, impact duration, and fracture length were
recorded. Pressure sensitive ﬁlm was used to record contact areas and pressures for each
impact. Fracture length was measured to the nearest millimeter. Impact force increased
with age for both rigid and compliant interfaces. Contact area for compliant interface
impacts increased with age but no change with age was noted for rigid interface impacts.
Bone ﬁ'actures initiated at bone-suture interfaces and not at the point of impact. Diastatic
fractures were observed on skulls as early as 4 days of age after impacts with a
compliant interface, while rigid interface impacts did not show suture damage until 10
days of age. Diastatic fractures were common for both interfaces on skulls aged 10 to 17

days, but disappeared after 17 days of age. For a given energy, impact of the skull with a

30

compliant interface caused more fracture damage than with a rigid interface for
specimens less than 17 days of age, but less damage for specimens aged 24 to 28 days.
The age-dependent response of the developing skull to impact with rigid and compliant
surfaces is important in forensic cases for determining potential differences between

testimony describing the events leading to injury and the observed fracture pattern.

31

 

INTRODUCTION

Head injuries account for 80% of fatal child abuse in younger children (Case et
al., 2001), while short falls rarely, if ever, cause serious injury or death (Reiber, 1993).
In one study of 89 children under 2 years of age, 19 of the 20 fatalities were due to abuse
(Hobb, 1984), while a study of 1,881 falls tom bed height reported no deaths (Belechri
et al., 2002). However, non-fatal injury ﬁ'om short falls (less than 3 feet) can still result
in skull ﬁacture especially for infants less than 12 months of age (Gruskin and
Schutzrnan, 1999). Due to the paucity of skull fracture tolerance data for infants and
young children, pediatric tauma involving single-event head injuries with related cranial
fractures represent one of the greatest challenges to forensic pathologists and
anthropologists.

Distinguishing between abuse and accidental tauma can be difﬁcult, as both may
produce similar types of injuries (Billmire and Myers, 1985). Speciﬁcally, linear,
complex, and depressed fractures can be seen in both types of cases (Reece and Sege,
2000; Wheeler and Shope, 1997). The most commonly fractured cranial bone in both
accidental injury and abuse cases is the parietal (Hobb, 1984; Meservy et al., 1987;
Leventhal et al., 1993). The risk of injury is also dependent on the contacting surface and
fall height (Bertocci et al., 2003; Chalmers et al., 1996). Other variables, such as the area
stuck, thickness of the skull, thickness of scalp and hair, and impact direction can also
affect the fracture pattern (Knight, 1991; Cooperman and Merten, 2001). To develop
reliable models for skull ﬁ'acture prediction, the effects of these aforementioned

variables must be better understood.

32

 

While several studies have documented important information regarding the
response of the infant skull and brain to blunt force impact, these data are limited to one
or two age groups. And because of ethical considerations, experimentation on human
pediatric specimens is very lirrrited. While scaling of the adult skull has met with some
success in predicting impact response of the pediatric skull (Prange et al., 2004), the
head of an infant is smaller and geometrically unlike that of an adult (Schneider et al.,
1986) and the validity of predicting skull fracture patterns in infants from adult data has
not been investigated. Using adult data to predict skull fracture patterns in the pediatric
skull may also be problematic due to the different stuctural (ease of deformation and
decreased threshold to ﬁacture) and mechanical properties of the infant skull (Thibault
and Margulies, 1998; Prange et al., 2000; Prange and Margulies, 2002). However, one
animal model of the human infant has begun to emerge regularly in the current literature.
A study by Dickerson and Dobbing (Dickerson and Dobbing, 1967) shows a similarity
in development of the cental nervous system when correlating months of the human to
weeks of the pig. Other studies have used porcine models to predict ﬁacture loads for the
infant human femur (Pierce et al., 2000) and stain in the braincase and sutures (Herring
and Teng, 2000). More recently, age-related rates of change in the bending rigidity of the
infant human and infant porcine skulls has been found to be similar for a correlation
based on months in humans to days in pigs (Baumer et al., in press).

The hypotheses of the current study were two-fold. First, impact energy required
to initiate a skull ﬁacture would depend on specimen age. Second, the extent of fracture
damage for a given energy impact would depend on impact interface. These data using a

developing porcine model may provide a baseline for the characterization of ﬁacture

33

patterns resulting from blunt force impacts onto different contact interfaces for the
developing human pediatric skull. These data may have ultimate utility in the study of
potential child abuse cases.

MATERIALS AND METHODS

Porcine specimens aged from 2 to 28 days (n=76) were obtained tom a local
supplier and stored at —20°C. All animals died of natural causes. The specimens were
ﬁozen within 12 hours of death. None of the specimens had outward signs of head
injury, which was later conﬁrmed during specimen preparation.

Each specimen was thawed at room temperature for 24 hours prior to testing. All
soft tissue was removed from the left side of the skull, leaving the periosteum to help
prevent drying of the bones and sutures. This side of each skull was placed in a bed of
air-hardening epoxy (Fibre Stand, Martin Senour Corp., Cleveland, OH). Capturing the
skull in this manner allowed deformation of the skull bones and sutures, but eliminated
gross motions (tanslations and rotations) of the skull as a whole. After the potting
material had hardened, the skull was bathed in phosphate-buffered saline (PB S) solution
in regular intervals until the experimental impact. Each skull, with the scalp in place,
was impacted in the center of the right parietal bone. The impact location was contolled
by a custom-made, 4 degree-of-ﬁ'eedom, lockable ﬁxture that was built to hold the
potted skull (Figure 3.1). The skull was oriented so that the impact was normal to the

bone surface.

34

Impact Head
Pressure Film

Potting

Positioning
Fixture

 

Figure 3.1. Skull positioning and set-up showing rigid impact head.

Each specimen was subjected to a single impact with a gravity-accelerated mass
(GAM) (Figure 3.2). A load tansducer (4.45 kN capacity, model AL311CV, Sensotec,
Columbus, OH) was mounted on the GAM behind the impact interface to record forces.
A single impact was assured by using an operational ampliﬁer comparator circuit to
monitor the impact forces and energize an electomagnetic solenoid to catch the
irnpactor immediately after the load returned to zero. Two interchangeable impact
interfaces were created for the GAM to represent rigid and compliant impact interfaces.
The rigid interface was a solid aluminum cylinder with approximately 16 cm2 of impact
surface. The compliant interface was a deformable aluminum material (1.10 MPa crush
strength Hexcel, Hexcel Corp., Stamford, CT), approximately 3 cm thick with a 16 cm2
impact surface, fastened to the rigid interface. Both interfaces had greater surface area

than the resulting contact area on the skull.

35

Stainless Steel Shaft

   
 
   
   
 
 
  

Vertical
Mounted Trolley
with Adjustable
Mass (GAM)

Impact Head

Skull with Parietal Bone
Exposed to Apparatus

Positioning
Fixture

  

Figure 3.2. Drop test ﬁxture set-up.

Input energy was adjusted by changing the drop height of the 1.67 kg GAM. A
second mass of 1.92 kg was necessary for specimens twenty-one days and older, as the
drop height was limited in the laboratory. The mass of the impact interface for each
impact was included in the total mass of the GAM, and the force data was checked for
inertial effects. Preliminary data collected at several ages was used to estimate the input
energy needed to initiate a ﬁacture using the rigid interface for each specimen age. In a
few cases (n=7), the ﬁrst impact with a rigid interface did not cause a fracture, so a
second, higher energy impact was performed. Impacts were always conducted in pairs.
All skulls impacted with the compliant interface were impacted at the same energy that
caused ﬁ’acture in the paired rigid impact on another skull of the same age. While the
input energy was the same for rigid and compliant impacts at each age, the energy of
impact required to cause fracture initiation had to be increased with specimen age. The

forces and impact durations were recorded at 10,000 Hz.

36

Pressure sensitive ﬁlm packets (Prescale, Fuji Film Ltd., Tokyo, Japan) were
attached with tape to the skull at the impact site to measure contact area and pressures
generated during impact. Low (0-10 MPa) and medium (10-50 MPa) range pressure
ﬁlms were stacked together and sealed between two sheets of polyethylene (0.04 mm
thick) to prevent exposure to ﬂuids during the impact (Atkinson et al, 1998). Using the
previously documented procedure, the ﬁlm was scanned and converted to pressure using
a dynamically calibrated density-to-pressure scale. Only the low pressure ﬁlm was
analyzed in these experiments as the pressures were not high enough to record on the
medium pressure ﬁlm.

Computed tomography (CT) scans were conducted for each skull after impact
(625 rim-thick coronal slices). The width of the skull vault and average thickness of the
parietal bone were obtained tom these scans using reconstuction software (Osirix
version 3.0, Open Source General Public License). The width of the vault was measured
at its widest point ﬁ'om parietal bone to parietal bone. Average parietal bone thickness
was determined from a slice near the center of the parietal bone.

Visual inspection was conducted on each skull after the soft tissue and
periosteum were removed from the impacted side. All suture damage and any visible
bone fractures were recorded. Each specimen was then cleaned via standard
anthropological procedures, removing all remaining soft tissue. Complete fracture
diagrams and measurements were made for each specimen. Total length of the skull
tactures was measured to the nearest millimeter with a ﬂexible measuring tape, which

conformed to the curvature of the skull.

37

The impact and pressure data were analyzed for age effects using linear
regression analyses. Comparisons between the interfaces were performed using a two-
factor (age, interface) AN OVA. Statistically signiﬁcant effects were reported for p<0.05.
RESULTS

Linear regression analyses showed a signiﬁcant increase (0.0516 mm/day) in
average parietal bone thickness with age (p<0.001), but no change in skull vault width
with age.

The input energy required to cause fracture increased tom 1.5 J at two days of
age to 11.5 J at twenty-eight days of age. A characteristic rapid drop in force appeared in
the force-time plot for all rigid interface impacts that were associated with ﬁacture
(Figure 3). In rigid interface impacts that did not cause ﬁacture, a more pronounced peak
was observed. Rapid drops in force during compliant interface impacts could not always

be associated with ﬁactures due to the deformation of the interface.

 

 

500 - N0
../

400 - mm"
2 300 ~
(D
8
LL 200 -

100 _ Fracture

o I f I
0 0.01 0.02 0.03
Time (see)

Figure 3.3. Overlay of force-time plots showing typical characteristics of an impact that
caused fracture and one that did not.

38

For both impact interfaces, linear regression analyses showed a signiﬁcant
increase in impact force with age that was required to produce a ﬁ'acture (p<0.001)

(Figure 3.4). The slopes and intercepts of impact forces versus age were not different

 

 

 

based on interface.
1600 -
E 1200 -
(D
2
1‘2 800 -
*5
S
g 400 -
o I l I
0 10 20 30

Age (days)
Figure 3.4. Peak force during impact for rigid and compliant interfaces.

The area of contact for the compliant interface signiﬁcantly increased with age

(p<0.001), but no signiﬁcant change was noted for the rigid interface (Figure 3.5).

 

1600 - 9 Rigid a Hexcell
0

NA

E1200 - a
3 ° °
5 800- , a
g ,.‘D' a

c 400 - a, ' n
8 e on Us

0

 

 

Age (daYS)

Figure 3.5. Contact area recorded on pressure sensitive ﬁlm during impacts for rigid and
compliant interfaces.

39

Bone fractures for all ages, and for both interfaces, occurred at sites away from
the point of impact (Figure 3.6). These bone fractures initiated at a suture-bone interface.
Frequently, bone fractures were accompanied by diastatic ﬁactures that were generally
located near the ﬁ'acture tip at the bone-suture interface. Diastatic ﬁactures caused by a
rigid interface impact were not present until specimens were ten days of age, while
diastatic fractures caused by a compliant interface were present as early as four days of
age (Figure 3.7). Diastatic fractures were common for both interfaces between ten and

seventeen days. Neither interface showed suture damage after seventeen days.

a b
i i I I} - Impact Site

  

Fracture
Figure 3.6. Representative fracture sites for (a) rigid and (b) compliant interface impacts
(ﬁve-day—old specimens shown). Fracture sites initiated at bone-suture boundaries.
Compliant interface impact showed increased fracture lengths at similar initiation sites
as the rigid interface impacts as well as diastatic fracture of coronal suture.

40

 

 

 

 

 

 

Ri'

100- _ _ I gld
DHexceIl

§

72’ 75—

8

.l-‘f

o 50- a

:5

S

(I)

.9

o 25-

o IIIIII II

 

 

 

 

 

 

 

I I I I I I I I I I I I

2 4 6 810121416182022242628
Age (days)

Figure 3.7. Frequency of diastatic fractures caused by both rigid and compliant
interfaces versus age.

A two-factor (age, interface) AN OVA performed on the total ﬁacture length
(bone and diastatic combined) revealed signiﬁcantly more damage caused by the
compliant interface than by the rigid interface on pig skulls less than seventeen days of
age, similar amounts of damage ﬁom both interfaces between seventeen and twenty-two
days of age, and signiﬁcantly more damage caused by the rigid interface than by the
compliant interface for specimens aged twenty-four to twenty-eight days (p<0.05)

(Figure 3.8).

41

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

150 - _ I Rigid
i , El Hexcell

E

g 100

:E

m

r:

c

-' i

E

13 50 -

E

u.

. -l l l
2 4 6 810121416182022242628
Age (days)

Figure 3.8. (Average) Fracture length generated by rigid and compliant impacts at the
same impact energy.
DISCUSSION

In an effort to better understand ﬁacture of the developing skull under known
conditions, the current study examined characteristics of impact induced ﬁacture to the
infant porcine skull with respect to age and impact interface. As would be expected, the
peak impact force increased with age. While an increase in the bending modulus of the
porcine skull with age would account for this increase in force with age, a previous study
shows that the bending modulus of infant porcine bone and bone-suture-bone specimens
under 4-point bending does not change signiﬁcantly with specimen age (Baumer et al.,
in press). Measurements of the average parietal bone thickness, however, showed a
signiﬁcant increase with age, which seemed to correlate with changes in the impact
force. It is expected that the increasing stength of the skull with age is due to changes in

geomety (thickness), and not due to an increase in bone quality with age.

42

There has been debate in the forensic community as to the mechanism of fracture
initiation on adult human skulls. Speciﬁcally, whether ﬁactures initiate at the point of
impact (Kroman, 2004) or at remote sites of “outbending” (Gurdjian et al., 1950). In the
current study, 70 of 76 infant porcine specimens displayed ﬁacture initiation at a suture-
bone boundary, and not at the center of the parietal bone (the impact location). The other
six specimens involved comminuted skull ﬁactures that obscured the site of initiation.
The locations of fracture initiation on infant porcine specimens in the current study
agreed with the ﬁndings of Gurdjian et al. (1950) for adult human skulls, however it is
currently unclear if the fracture initiation sites observed in this study are due to the
immature nature of the cranial bone, differences in porcine cranial geomety or a
combination these factors. Despite potential differences limiting direct comparisons with
adult human fracture information, the infant porcine model developed in the current
study showed that ﬁacture initiation occurred in relatively predictable locations at bone-
suture interfaces for impacts on the parietal bone against a ﬂat surface. This information
may be expanded in the future to other interface conditions and should be correlated
with ﬁacture patterns in the human infant skull from case ﬁles. These ﬁndings suggest
that skull fractures on the developing skull from impacts with rigid or compliant ﬂat
surfaces may be located remote tom the site of the impacting force and nearer to the
cranial sutures.

The current study produced a greater amount of ﬁacture on younger infant
porcine skulls using a compliant than rigid interface. The concentrated force of an
impact with a rigid interface was expected to produce a larger amount of bone ﬁacture

than an impact with a compliant interface where the same force intensity would be

43

distributed over a larger contact area and contact stesses would be reduced. This study,
however, showed that the relative amount of fracture between these interfaces was
dependent on age. For infant pigs less than seventeen days of age, a compliant interface
caused signiﬁcantly more damage than a rigid interface for the same impact energy,
while for specimens twenty-four days of age and older the rigid interface generated more
damage than the compliant interface. Specimens younger than seventeen days of age
also showed ﬁequent damage to the sutures, while specimens older than seventeen days
of age had bone fractures alone. In a previous study on beam specimens cut ﬁ'om infant
porcine skulls and loaded in 4-point bending, it has been shown that there is a signiﬁcant
difference in the relationship between bone and bone-suture-bone material properties for
specimens aged fourteen and eighteen days (Baumer et al., in press). These changes
resulted in a signiﬁcant increase in the relative stength of the coronal suture at eighteen
days to values more similar to that of bone. This effect correlates well with the
disappearance of suture damage and a change in interface-dependent ﬁacture differences
documented in the current study. The age-dependent response of the developing skull to
impact with rigid and compliant surfaces is important in forensic cases for determining
potential differences between testimony describing the events leading to injury and the
observed fracture pattern. Speciﬁcally, the infant porcine model suggests that very
young skulls may actually be more susceptible to extensive skull ﬁacturing tom an
impact against a compliant surface than ﬁom an impact against a more rigid surface.
Further investigations will be needed to create a more comprehensive set of patterns for

various impact interfaces and energies.

In an effort to explain the interface- and age-dependent results of the current
study, it was noted that the contact area for the compliant interface impacts was
generally greater than that for the rigid interface impacts. Yet, the generated impact force
was not different between interfaces. It was hypothesized that the increased contact area
for the compliant interface impacts resulted in a higher state of stess close to the
boundaries of the parietal bone (sutures). It may then be suspected that the mechanical
properties of the sutures were more critical during impacts with a compliant than with a
rigid interface. For specimens under seventeen days of age, when the sutures are
signiﬁcantly more compliant than bone, large deformations at the sutures during impact
for both interfaces may have generated high levels of stess near the boundaries of the
parietal bone at the sutures. Therefore the risk of damage at the suture may have
increased, resulting in the high frequency of diastatic ﬁactures. Furthermore, the higher
state of peripheral stesses caused by the compliant interface may have promoted
fracture propagation, resulting in the increased fracture length documented for the
compliant interface versus the rigid interface. The increase in suture stength that occurs
between fourteen and eighteen days of age may have allowed a better distibution of
impact stesses across the boundaries of the parietal bone, resulting in lower overall
levels of stess in the parietal bone and across suture lines. This effect may help explain
why in the relatively older infants the compliant interface actually generated less
ﬁacturing to the skull than for the rigid interface, since the rigid interface would be
expected to generate more localized contact stesses nearer the site of impact.

In summary, this study documented an increase in impact energy required to

initiate skull ﬁactures as specimen age increased. Furthermore, there was relatively more

45

fracture damage generated for a compliant than a rigid interface with equal input energy
at the younger infant ages. While a direct correlation with infant human ﬁacture patterns
is not currently available ﬁ'om case ﬁles, the experimental observations from the
developing porcine model may aid in understanding some of the limited data that is
available today. The experimental documentation of ﬁacture initiation energy and the
extent of skull fracture with respect to interface as a ﬁmction of specimen age may be of

critical importance in ﬁxture investigations of potential childhood abuse.

46

REFERENCES

Baumer T, Passalacqua N, Powell B, Fenton T, Haut R., Accepted for Publication, “Age
Dependent Mechanical Properties of the Infant Porcine Parietal Bone and a Correlation
to the Human,” Journal of Biomechanical Engineering, in press.

Belechri M, Petridou E, Trichopoulos D., 2002, “Bunk versus conventional beds: a
comparative assessment of fall injury risk,” J oumal of Epidemiology and Community
Health, 56, pp. 413-7.

Bertocci G, Pierce M, Deemer E, Aguel F, Janosky J, Vogeley E., 2003, “Using Test
Dummy Experiments to Investigate Pediatic Injury Risk in Simulated Short-Distance
Falls,” Archives of Pediatric Adolescent Medicine, 157(5), pp. 480-6.

Billmire M. and Myers P., 1985, “Serious head injury in infants: accident or abuse?,”
The American Academy of Pediatrics, 75, pp. 340-2.

Case M, Graham M, Handy T, J entzen J, Monteleone J ., 2001, “Position Paper on Fatal
Abusive Head Injuries in Infants and Young Children,” The American Journal of
Forensic Medicine and Pathology, 22, pp. 112-22.

Chalmers D, Marshall S, Langley J, Evans M, Brunton C, Kelly A, et al., 1996, “Height
and surfacing as risk factors for injury in falls ﬁ'om playground equipment: a case-
contol study,” Injury Prevention, 2, pp. 98-104

Cooperman D. and Merten D., 2001, “Skeletal Manifestation of Child Abuse,” In: Child
Abuse: Medical Diagnosis and Management. Philadelphia: Lippincott, Williams, and
Wilkins, pp. 123-56.

Dickerson J. and Dobbing J ., 1967, “Prenatal and postratal growth and development of
the cental nervous system of the pig,” Proceedings of the Royal Society of London,
166(1005), pp. 384-95.

Gruskin K and Schutzrnan S., 1999, “Head Trauma in Children Younger Than 2 Years,”
Archives of Pediatric Adolescent Medicine, 153, pp. 1 5-21.

Gurdjian E, Webster J, Lissner H., 1950, “The Mechanism of Skull Fracture,”
Radiology, 54(3), pp. 313-39.

Herring S. and Teng S., 2000, “Stain in the Braincase and Its Sutures During Function,”
American Journal of Physical Anthropology, 112, pp. 575-93.

Hobb C., 1984, “Skull Fracture and the Diagnosis of Abuse,” Archives of Disease in
Childhood, 59, pp. 246-52.

47

Knight 3., 1991, “Fatal child abuse,” In: Forensic Pathology. London: Edward Arnold,
pp. 457-73.

Kroman A., 2004, “Experimental Study of Fracture Propagation in the Human Skull: A
Re-Testing of Popular Theories,” Proceedings of the American Academy of Forensic
Sciences, Dallas, TX, H79:314.

Leventhal J, Thomas S, Rosenﬁeld N, Markowitz R., 1993, “Fractures in young
children: distinguishing child abuse from unintentional injuries,” American Journal of
Diseases of Children, 147, pp. 87-92.

Meservy C, Towbin R, McLaurin R, Myers P, Ball W., 1987, “Radiographic
characteristics of skull ﬁactures resulting ﬁ'om child abuse,” American Journal of
Roentgenology, 149, pp. 173-5.

Pierce M, Valdevit A, Anderson L, Inoue N, and Hauser D., 2000, “Biomechanical
Evaluation of Dual-Energy X-Ray Absorptiomety for Predicting Fracture Loads of the
Infant Femur for Injury Investigation: An In Vito Porcine Model,” Journal of
Orthopaedic Trauma, 14(8), pp. 571-6.

Prange M, Luck J, Dibb A, Van Ee C, Nightingale R, Myers 3., 2004, “Mechanical
Properties and Anthropomety of the Human Infant Head,” Stapp Car Crash Journal, 48,
pp. 279-99.

Prange M. and Margulies S., 2002, “Regional, Directional, and Age-Dependant
Properties of the Brain Undergoing Large Deformations,” Journal of Biomechanical
Engineering, 124(2), pp. 244-52.

Prange M, Meaney D, Margulies S., 2000, “Deﬁning brain mechanical properties:
Effects of region, direction, and species,” Stapp Car Crash Journal, 44, pp. 205-13.

Reece R and Sege R., 2000, “Childhood Head Injuries: Accidental or Inﬂicted,”
Archives of Pediatic Adolescent Medicine, 154, pp. 11-15.

Reiber G., 1993, “Fatal falls in childhood: How Far Must Children Fall to Sustain Fatal
Head Injury?,” American Journal of Forensic Medicine and Pathology, 14, p. 201-7.

Schneider L, Lehman R, Pﬂug M, Owings C., 1986, “Size and Shape of the Head and
Neck From Birth to Four Years,” Washington, DC. The Consumer Product Safety
Commission, Report No.: UMTRI-86-2.

Thibault K. and Margulies S., 1998, “Age-Dependant Material Properties of the Porcine

Cerebrum: Effect on Pediatic Head Injury Criteria,” Journal of Biomechanics, 31(12),
pp. 1 1 19-26.

48

Wheeler D. and Shope T., 1997, “Depressed skull fracture in a 7-month old who fell
from bed,” Pediatrics, 100, pp. 1033-4.

49

CHAPTER FOUR

CHARACTERISTICS AND PREDICTION OF CRANIAL CRUSH
INJURIES IN CHILDREN

ABSTRACT

Experimentally applied bilateral crushing forces on adult cadaver skulls and even a few
case reports on children have shown similarities in basilar ﬁacture locations. It has even
been suggested that the overall path of tacture may be related to the location of applied
force on the skull. This study documents 4 clinical cases of fatal crush injuries to young
children between 1.5 and 6 years of age with correlations between modeled stess and
clinically observed fracture patterns. The damage in the clinical cases was concentated
in the basicranium. Each of the cases presented fracture patterns that bridged the impact
sites and traversed the middle cranial fossa in the area of the spheno-occipital
synchondrosis. The crushing forces observed in the clinical cases were recreated on a
simpliﬁed ﬁnite element model of a cranium by applying bilateral pressures to regions
correlating to the speciﬁc exterior injury locations of each case. Numerous trials were
run within the elastic range of bone to develop a representative and clearly characterized
pattern of principal stess directions. Areas of tensile stess were assumed to be the most
likely location(s) of potential fracture. The regions of largest tensile stess developed on
the skull were compared to the clinically observed fracture patterns in the 4 cases,
revealing that locations of high tensile stess in all four cases were located on the
basicranium with low levels of stess on the cranial vault. Additionally, regions of high
tensile stess appeared to follow the ﬁacture path(s) extending tom the location of

maximum stess. These results suggest that pre-failure stess ﬁeld diagrams may give

50

some insight into the projected fracture propagation paths, although prediction will not
be exact. Also, these analyses of the modeled stesses and the clinically observed
fractures suggest that quasi-static bilateral loading of the cranium may lead to

predictable fracture of the basicranium, possibly independent of age.

51

INTRODUCTION

Fracture interpretation is a necessary component of forensic analysis and
contributes to the determination of the cause and manner of death. Skeletonized remains,
in particular, lack skin and scalp abnormalities that enhance the complete description of
a head injury or injuries. Thus, a clear understanding of fracture propagation becomes
vital to the forensic professional. Head injuries are often the result of dynamic forces in
which a moving object strikes the cranium at a considerable speed or the cranium makes
contact with a static object that causes it to rapidly decelerate (Duhaime et al., 1995). In
cases of crushing injuries, however, a static load is applied relatively slowly to the
cranium, initially causing deformation with eventual ﬁacturing if pressure continues or
increases. In experimental tests on the effects of bilateral crushing forces on cadaver
skulls, Russell and Schiller (1949) report that avulsion of the petous portion, fracture
through the dorsum sellae, and principal fracture lines running in the direction of the
compressive forces were common features observed.

There are several case reports of basilar skull damage in the literature involving
both crushing and impulse based loading scenarios mainly focusing on adult crania. A
study of 15 cases of crushing injuries to adult skulls aged 14-63 years and
experimentally induced crushing fractures on 11 cadaver crania revealed similarities in
basilar fracture patterns (Russell and Schiller, 1949). However, a study by Harvey and
Jones (1979) claimed that fractures through both petous bones was not an indicator of
impact site. Another study by Yasue (1981) then suggested that the overall path of
ﬁacture through the basicranium could, in fact, be related to the region of the skull

impacted. More recently, a study by Kroman (2005) attempted to produce cranial base

52

(hinge) fractures in two unembalmed cadaver adult human crania specimens using a
drop tower. Hi gh-speed video dernonstated that ﬁ'actures initiated at the site of impact
and propagated through the middle cranial fossa, with some additional minor fractures
surrounding the forarnen magnum. In this study the vaults had been removed so it is
unclear if the ﬁactures would have terminated at the points of applied forces (ﬁxture and
impactor). Kroman further notes a very wide separation of the failing bones during
impact which no doubt would cause very severe soft-tissue damage in the cranial base;
this correlates to the high amount of energy and lethality involved in these scenarios.

It is assumed that fractures will generally follow the path of least resistance, but
this path may be different between the child and adult cranium. While all of the
aforementioned studies focused on adult crania, some recent case reports on quasi-static
loading to the crania of children have also presented observations of basilar ﬁactures
(T akeshi et al., 2005; Duhaime et al., 1995). The tend of quasi-static loading producing
basilar tactures in both adults and children may suggest that prediction of these
fractures is possible.

The Finite Element (FE) method is a powerful analytical modeling tool that can
be used to determine internal stesses and stains within a complex solid. Several studies
have shown that maximum principal stess (or stain) directions from a modeled
stucture correlate well with areas of experimentally or clinically observed bone
ﬁactures (Graham et al., 2000; Silva et al., 1998; Doorly and Gilchrist, 2006). While
modeling of ﬁacture propagation generally involves high computational costs, one
theory suggests that there is a close, although not exact, relationship between pre-failure

stess tajectories and the ultimate path of propagation. In brief, Frank and Lawn (1967)

53

proposed that crack growth occurs in such a way as to maximize the energy dissipation
per unit length of growth, and that the greatest amount of dissipation occurs when the
crack plane is perpendicular to the direction of greatest principal tensile stess. In other
words, the directions of principal tensile stess developed on a solid may be a reasonable
predictor of propagation direction.

In this study, four cases of childhood fatalities from crushing head injuries
inﬂicted by vehicle tires exhibited similar fracture patterns despite differences in age and
extent of force. This raised the question as to whether certain fracture patterns are
indicative of bilateral crushing injuries and if these could be predicted based on the type
of loading and the biomechanical properties of the skull.

MATERIALS AND METHODS

Four clinical cases of fatal crush injuries to young children between 1.5 and 6
years of age under known conditions were used as a baseline for applied forces and
resulting ﬁacture patterns. In all cases the skull was tapped between a vehicle tire and
the ground. The contact sites and skull fractures were identiﬁed and diagramed during
autopsy by the investigating forensic pathologist (M.N.). Biomechanical modeling was
then conducted in an attempt to recreate and explain the similarities exhibited in the
fracture patterns of these cases.

A simpliﬁed model of a skull was created in a ﬁnite element package (Abaqus
v.6.3, Hibbitt, Karlsson & Sorensen, Inc., Pawtucket, RI, USA) (Figure 4.1). This model
consisted of synmrety about the sagittal plane, but not about the coronal or tansverse
planes. Each major bone of the vault was modeled separately, assumed to have a

uniform thickness, and constained at the suture joints to form the ﬁll] cranial model.

54

The base of the cranium was modeled as a ﬂat surface with the same thickness as the
vault. The primary landmarks in the basal region were modeled: a hole was added to
simulate the foramen magnum, dense oblong extusions lateral to the foramen magnum
were generated to represent the petous portions, and anterior to the foramen magnum

the thickness was reduced to represent the region of the spheno-occipital synchondrosis.

 

 

 

 

 

 

Foramen
Petous / C\ magnum
portions *9
Posterior N m
Spheno-occipital
synchondrosis

 

 

 

 

 

Figure 4.1. Simpliﬁed cranial model showing modeled landmarks of the basicranium.
View of endocranial surface with vault removed.

FE analysis was utilized in an effort to reproduce and examine possible
relationships between the theoretical pre-failure stesses developed on the skull and the
fracture patterns observed clinically. The magnitude of the force applied to the cranial
model was the same in every case and was scaled down such that the modeled cranium
would not actually fail. The crushing forces reported in the clinical cases were recreated
by applying the force over a region of the cranium correlating to the speciﬁc exterior
injury locations for each case. Numerous trials were run within the elastic range of bone

to develop a representative and clearly characterized pattern of principal stess and stain

55

directions. Areas of tensile (principal) stess were assumed to be the most likely
location(s) of potential tacture.

RESULTS

CASE REPORTS

Case I
A 5-year-old female child was stuck by her family’s sedan when she fell in an

attempt to stop the vehicle after her brother accidentally engaged the tansrrrission. Her
head was run over by the right front tire of the car. The autopsy revealed an intact
cranial vault with a cranial base fracture that taversed the lateral right anterior cranial
fossa, the midline of the middle cranial fossa, and the left lateral aspect of the posterior
cranial fossa. This fracture bridged the blunt impact sites on the right anterior and left
posterior aspects of the head.

Case 2
An unhelmeted 6-year-old male was stuck by a car while riding his bike. His

head was run over by the vehicle’s left rear tire. Autopsy revealed an intact cranial vault.
The cranial base, however, was bisected by a gaping X-shaped tacture centered in the
midline of the middle cranial fossa. There were two adjacent arms of the ﬁacture in the
right middle cranial fossa. The other two adjacent arms were in the posterior aspect of
the left middle cranial fossa and the anterior aspect of the left posterior cranial fossa.
This ﬁacture bridged the blunt impact sites on the right anterior and lateral aspects of the
head and the left lateral and posterolateral aspects of the head.

Case 3
A 3-year-old male was run over attempting to run away from a farm grain wagon

that had begun to roll. His head was run over by one of the wheels of the wagon.

Autopsy revealed gaping linear fractures that taversed the parietal bones and the right

56

and left sides of the ﬁontal bone. One end of each fracture intersected the widely gaping
right side of the coronal suture, other ends were continuous with fractures of the cranial
base. A dominant cranial base fracture taversed the left petous bone, the midline of the
middle cranial fossa, the anterior aspect of the right middle cranial fossa and the
posterolateral aspect of the right anterior cranial fossa. Additional intersecting fractures
were centered in the left middle and posterior cranial fossae.

Case 4
A 1.5-year-old female was entapped under a small moving van and her head was

run over by the right rear wheel. Autopsy revealed an intact cranial vault but extensive
intersecting fractures of the cranial base that predominately taversed the middle and
posterior cranial fossae ﬁ'om side to side. A large fracture also bisected the anterior
cranial fossae.
Fracture Pattern Similarities

Each of the cases presented tacture patterns that bridge the impact sites and
taversed the middle cranial fossa in the area of the spheno-occipital synchondrosis.
Additionally, the damage was concentated in the basicranium leaving the cranial vault
intact in all cases except the 3-year-old boy impacted by the farm grain wagon.
Furthermore, general amounts of fracture increased with inferred increase in magnitude
of applied force (grain-wagon vs. car) or decrease in age.

The four cases presented three contact scenarios with four different fracture

patterns. Speciﬁcally, the contact forces in Case 1 were applied primarily to the frontal
and occipi tal bones in a direction between anterior-posterior and lateral (Figure 4.2a),

the contact forces in Cases 2 and 3 were also applied to the ﬁontal and occipital bones

57

but had an additional lateral component (Figure 4.2b, c), and the contact forces in Case 4

were applied to the lateral surfaces of the cranial model (Figure 4.2d).

 

Figure 4.2. Diagrams of fractures due to quasi-static crushing forces to skulls aged (a) 5
years — Case 1, (b) 6 years — Case 2, (c) 3 years — Case 3, and (d) 1.5 years — Case 4.
White cranial margins represent contact of applied forces (ground and tire).

FE Analysis

Modeling of all four cases revealed that locations of high tensile stess and high
tensile stain were located on the basicranium. There were some areas of lower tensile
stess at the sites of the applied load and in a few areas on the cranial vault, but due to
the assumption that failure would occur at locations of high stess, these areas were
ignored for this analysis. The maximum tensile stess for all models were similar
(standard deviation of :I:5%). The tensile stesses developed on the skull were ﬁltered to
display only the highest 20% for correlation to fracture patterns. The leng th of the
arrow at each location represented the relative magnitude of stess at that location.

Case 1 presented the simplest ﬁacture pattern as a single linear fracture
taversing the basicranium from the right greater wing of the sphenoid across the
spheno-occipital synchondrosis to the left lateral portion of the occipital, connecting the
loading sites (Figure 4.3a). An overlay of the directions of maximum principal tensile

stesses revealed a linear concentation on the right side of sphenoid and in the area of

58

the spheno-occipital synchondrosis, which correlated well with sections of the observed

fracture (Figure 4.3b).

   

Figure 4.3. (Case 1) Crushing injury to the skull of a 5-year-old. Diagrams of (a)
clinically observed ﬁ’actures and (b) an overlay of the highest 20% of principal tensile
stesses produced from a FE model. White cranial margins represent contact of applied
forces (ground and tire).

Cases 2 and 3 presented slightly more complex fracture patterns. In part, the
complexity was increased due to the similarities in loading but dissimilarities in fracture
(Figure 4.4a and Figure 4.5a, respectively). A similar ﬁacture in both cases stetched
from the right aspect of the sphenoid through the sella turcica and along the superior
aspect of the petous portion of the temporal. The overlay of the directions of maximum
principal tensile stesses was identical in both cases due to the similarity of loading
patterns. There was correlation between the stess and the ﬁacture for both cases,

especially along the line of similar ﬁacture, although each had areas of high stess where

fracture did not occur (Figure 4.4b and Figure 4.5b).

59

 

Figure 4.4. (Case 2) Crushing injury to the skull of a 6-year-old. Diagrams of (a)
clinically observed fractures and (b) an overlay of the highest 20% of principal tensile
stesses produced from a FE model. White cranial margins represent contact of applied
forces (ground and tire).

 

Figure 4.5. (Case 3) Crushing injury to the skull of a 3-year-old. Diagrams of (a)
clinically observed fractures and (b) an overlay of the highest 20% of principal tensile
stesses produced from a FE model. White cranial margins represent contact of applied
forces (ground and tire).

Case 4 presented the most complex fracture pattern extending both longitudinally
and tansversely through the basicranium (Figure 4.6a). Fractures ran longitudinally
across the midline of the anterior cranial fossa, and horizontally through the middle and
posterior cranial fossa with various areas of comminution. The overlay of the directions

of maximum principal tensile stesses revealed an area of complex stess in the same

area as the highly comminuted fracture of the right side of the cranial base. Additionally,

60

a more linear pattern of stesses occurred near the ﬁactures on the left side of the skull

(Figure 4.6b).

 

Figure 4.6. (Case 4) Crushing injury to the skull of a 1.5 year old. Diagrams of (a)
clinically observed ﬁactures and (b) an overlay of the highest 20% of principal tensile
stesses produced from a FE model. White cranial margins represent contact of applied
forces (ground and tire).

DISCUSSION

The analyses performed in this study represent a quantitative, theoretical attempt
to correlate stess and stain intensities and their distributions to fracture patterns on the
skull caused by quasi-static bilateral pressures. A linear, small stain, isotopic, uniform
thickness model of a human calvarium was constructed to maintain some manageability
in the preliminary modeling. Analyses of the modeled stesses and the clinically
observed fractures suggested that quasi-static bilateral loading of the cranium may lead
to predictable fracture of the basicranium, although the predictions were not exact. The
promising nature of these results suggest that it may be useful to develop a more
complex model for a more accurate prediction of fractures on a case-by-case basis.

The location of maximum tensile stess would be expected to be the site of
fracture initiation, and the location of maximum stess in each of the four models

correlated well with the area(s) of ﬁacture in each of the clinical cases. However, Frank

61

and Lawn (1967) also suggest that paths of high tensile stess may approximate the path
of ﬁ‘acture propagation. In the four cases presented in this study, regions of high tensile
stess appeared to follow the path of the fracture extending from the location of
maximum stess. Fracture propagation, however, requires increasing levels of stess to
overcome the dissipation of energy that occurs with the previous ﬁacturing. These
increasing levels of stess may ultimately result in additional ﬁacture sites on the
material surface (Paterson and Wong, 2005). Two or more isolated regions of the skull
developed high tensile stesses in Cases 2-4, which may have resulted in multiple sites of
fracture initiation with propagation extending from each site. Multiple initiation sites
with subsequent propagation joining the sites could explain the complexity of the
ﬁactures seen in Cases 2-4, as opposed to the singular region of ﬁacture and stess
intensity developed in Case 1. The four cases presented in this study suggest that pre-
failure stess ﬁeld diagrams may, in fact, give some insight into the projected ﬁacture
propagation paths and, ultimately, fracture patterns developed during quasi-static
bilateral loading of the cranium.

The cases and analyses presented in this study supported some general tends for
quasi-static bilateral loading scenarios. Autopsy revealed tactures through the middle
cranial fossa in the region of the spheno-occipital synchondrosis in all cases. Modeling
displayed high tensile stesses through this region in Cases 1-3; with Cases 2 and 3
developing maximum levels of stess at this landmark. However, modeling of Case 4
showed no region of high tensile stess passing through the spheno-occipital
synchondrosis. Interestingly, Case 4 involved the youngest child and one of the most

massive vehicles (moving van). This result may suggest that the material model should

62

be altered as a ﬁmction of age, especially for the very young pediatic patient. Autopsies
also revealed ﬁactures connecting the loading sites through the basicranium in all four
cases. The models showed regions of high tensile stess in this region for Cases 1-3,
although the tensile stesses in Case 4 did not seem to follow the same pattern. Intact
cranial vaults were also present in three of four cases. The models displayed no
signiﬁcant levels of stess on the vault that would be expected to cause ﬁacture. It is
possible that the mass of the vehicle (grain wagon) caused excessive fracture
propagation that was not able to be modeled in Case 3. All of the ﬁacture related
observations seen in the current study showed similarities to experimentally developed
ﬁactures on adult cadaver skulls (Russell and Schiller, 1949) suggesting that these
patterns of quasi-static loading of the cranium may be somewhat age independent,
except for the very young patient; likely as a function of cranial geomety which would
remain essentially the same throughout life.

In summary, a simple model of a cranium subjected to clinically observed
bilateral crushing forces was largely able to predict the characteristics of the developed
ﬁ'acture patterns. The results of this study were not meant to propose that the theoretical
stesses developed on this simpliﬁed cranial model actually predict an arbitary fracture
pattern developed clinically. These results do show that a simpliﬁed analysis of tensile
stesses developed tom some basic information on boundary loads and locations on the
cranium can generate patterns of fracture that are consistent with clinical fractures. The
four cases in conjunction with the FE model suggest that craniums subjected to bilateral

crushing forces tend to fracture in predictable ways. Further, the regions of high tensile

63

stess that develop between the locations of loading on the pediatric cranium coincide
with the locations of tensile fracture.

The predictability of this pediatric cranial ﬁ'acture pattern has practical
application. Medical examiner and coroner death investigations attempt to establish
circumstances of injury. In cases of unknown or questionable circumstances, this
ﬁacture pattern would suggest a particular mechanism of injury (relatively slow
application of a static load) consistent with a heavy wheeled vehicle. The described
ﬁacture pattern would be less consistent with other, more common injuries such as those
sustained in motor vehicle collisions (ejection or internal tumbling) and in the setting of
inﬂicted blunt tauma. This information may prove relevant in clarifying unexplained
perimortem events in cases of crushing injuries to the pediatric cranium, particularly if

information from examination of skin and scalp is not possible due to skeletonization.

64

REFERENCES

Doorly M, Gilchrist M., 2006, “The use of accident reconstuction for the analysis of
taurnatic brain injury due to head impacts arising from falls,” Computer Methods in
Biomechanics and Biomedical Engineering, 9(6), pp. 371-7.

Duhaime A-C, Eppley M, Margulies S, Heher K, Bartlett S., 1995, “Crush Injuries to the
Head in Children,” Neurosurgery, 37(3), pp. 401 -7.

Frank F, Lawn B., 1967, “On the Theory of Hertzian Fracture,” Proceedings of the
Royal Society of London, 299(1458), pp. 291-306.

Graham R, Oberlander E, Stewart J, Grifﬁths D., 2000, “Validation and use of a ﬁnite
element model of C-2 for deterrrrination of stess and ﬁacture patterns of anterior
odontoid loads,” Journal of Neurosurgery (Spine 1), 93, pp. 117-25.

Harvey F, Jones A., 1980, “‘Typical’ Basal Skull Fracture of Both Petous Bones: An
Unreliable Indicator of Head Impact Site,” Journal of Forensic Sciences, 25(2), pp. 280-
6.

Paterson M, Wong T., 2005, “Grifﬁth Theory of Brittle Failure,” In: Experimental Rock
Deformation — The Brittle Field. Berlin: Springer, pp. 45-57.

Russell W, Schiller F ., 1949, “Crushing Injuries to the Skull: Clinical and Experimental
Observations,” Journal of Neurology, Neurosurgery, and Psychiaty, 12, pp. 52-60.

Silva M, Keaveny T, Hayes W., 1998, “Computed Tomography-Based Finite Element
Analysis Predicts Failure Loads and Fracture Patterns for Vertebral Sections,” Journal of
Orthopaedic Research, 16, pp. 300-8.

Takeshi M, Okuchi K, Nishiguchi T, Seki T, Watanahe T, Ito S, et al., 2006, “Clinical
Analysis of Seven Patients of Crushing Head Injury,” The Journal of Trauma-Injury,
Infection, and Critical Care, 60(6), pp. 1245-9.

Tortosa J, Martinez-Lage J, Poza M., 2004, “Bitemporal Head Crush Injuries: Clinical

and Radiological Features of a Distinctive Type of Head Injury,” Journal of
Neurosurgery, 100, pp. 645-51.

Yasue M., 1981, “Clinical and experimental investigation of mediobasal skull ﬁacture,”
Neurol Med Chir (Tokyo), 21, pp. 1041-9.

65

CHAPTER FIVE

FINITE ELEMENT MODELING OF RIGID AND COMPLIANT
SURFACE IMPACTS TO THE PEDIATRIC PORCINE PARIETAL
BONE

ABSTRACT

Head injuries account for 80% of fatal child abuse in younger children. Yet, in many
cases the ability to determine child abuse from accidental injury may be difﬁcult. Head
injuries may occur when a moving object stikes the cranium at a considerable speed or
when the cranium strikes a static object that causes the skull to rapidly decelerate. This
study documents theoretical principal stess and stain directions developed on a cranial
model with respect to age and interface in an effort to identify ﬁacture characteristics for
the developing skull under known conditions. The impact conditions generated
experimentally for rigid and compliant interface impacts to porcine skulls of various
ages were recreated on a simpliﬁed ﬁnite element model. The model accounted for
changes in bone thickness as well as material properties of bone and suture with age.
Numerous trials were run within the elastic range of the materials to develop a
representative and clearly characterized pattern of pre-failure principal stess and stain
directions. Maximum principal stess and stain directions revealed distinct locations of
expected fracture regardless of age or interface. However, the extent and magnitude of
stains at any given location were dependent on both age and interface. Experimental
results showed that blunt force tauma to the parietal bone consistently resulted in
ﬁacture at the bone-suture boundaries of the parietal bone, which correlated well with
areas of high stain. Young specimens showed high levels of stain in the sutures while

old specimens showed limited frequency and magnitude of stain within the sutures.

66

These ﬁndings correlated well with previous ﬁndings of signiﬁcantly higher rates of
suture failure in young specimens than in old specimens. However, suture damage only
occurred experimentally in 29% of younger specimens, while bone tacture occurred in
97%. Due to the relative differences in the elastic moduli of bone and suture it was
expected that similar levels of stain could cause failure of the bone prior to failure of the
sutures. Interestingly, the larger contact area generated during compliant interface
impacts also seemed to produce stains within the bone distributed nearer the parietal
bone boundaries. This distribution of high stains may have promoted propagation of
fractures initiating at the boundaries, resulting in the greater amount of observed damage

on the cranium produced from compliant interface impacts.

67

INTRODUCTION

Skull fracture interpretation is a necessary component of forensic analysis and
contributes to the determination of the manner and cause of death. Head injuries are
often the result of dynamic forces in which a moving object strikes the cranium at a
considerable speed or the cranium makes contact with a static object that causes the skull
to rapidly decelerate (Duhaime et al., 1995). Head injuries account for 80% of fatal child
abuse in younger children (Case et al., 2001). Yet similar types of injuries can be
generated by abusive and accidental tauma (Billmire and Myers, 1985) and, due to the
paucity of cranial fracture tolerance data for infants and young children, pediatric tauma
involving single-event head injuries with related cranial ﬁ'actures represent one of the
greatest challenges to forensic pathologists and anthropologists. Thus, a clear
understanding of fracture patterns becomes vital to the forensic professional.

While several studies have documented important information regarding the
response of the infant cranium and brain to blunt force impact, these data are limited to
one or two age groups. And because of ethical considerations, experimentation on
human pediatric specimens is very limited. Scaling of the adult cranium has met with
some success in predicting impact response of the pediatic cranium (Prange et al., 2004)
but, the head of an infant is smaller and geometrically unlike that of an adult (Schneider
et al., 1986) and the validity of predicting cranial ﬁacture patterns in infants from adult
data has not been investigated. Using adult data to predict fracture patterns in the
pediatric cranium may also be problematic due to the different stuctural (ease of
deformation and decreased threshold to ﬁacture) and mechanical properties of the infant

cranium (Thibault and Margulies, 1998; Prange et al., 2000; Prange and Margulies,

68

2002). However, one animal model of the human infant has begun to emerge regularly in
the current literature. A study by Dickerson and Dobbing (1967) shows a similarity in
development of the cental nervous system when correlating months of the human to
weeks of the pig. Other studies have used porcine models to predict fracture loads for the
infant human femur (Pierce et al., 2000) and strain in the braincase and sutures (Herring
and Teng, 2000). More recently, age-related rates of change in the bending rigidity of the
infant human and infant porcine craniums has been found to be similar for a correlation
based on months in humans to days in pigs (Baumer et al., in press).

The Finite Element (FE) method is a powerful analytical modeling tool that can
be used to determine internal stesses and stains within a complex solid. Several studies
have shown that maximum principal stess (or stain) directions ﬁ'orn a modeled
stucture correlate well with areas of experimentally or clinically observed bone
ﬁactures (Graham et al., 2000; Silva et al., 1998; Doorly and Gilchrist, 2006). While
modeling of fracture propagation generally involves high computational costs, one
theory suggests that there is a close, although not exact, relationship between pre—failure
stess/stain tajectories and the ultimate path of propagation. In brief, Frank and Lawn
(1967) propose that crack growth occurs in such a way as to maximize the energy
dissipation per unit length of growth, and that the greatest amount of dissipation occurs
when the crack plane is perpendicular to the direction of greatest principal tensile stess.
In other words, the directions of principal tensile stess developed on a solid may be a
reasonable predictor of the propagation path.

In this study, blunt impact tauma to the center of the parietal bone was modeled

for comparison to ﬁacture patterns experimentally generated on the developing porcine

69

cranium. The hypothesis was that age and interface characteristics of a given impact
would deﬁne the stess and stain ﬁelds developed on the cranium, leading to predictable
fracture patterns. A secondary goal was to examine possible stess or stain related
causes for the experimentally generated fractures that consistently occurred away from
the impact site at the bone-suture boundaries (Chapter 3).
MATERIALS AND METHODS

A simpliﬁed model of a porcine cranium was created in a ﬁnite element package
(Abaqus v.6.3, Hibbitt, Karlsson & Sorensen, Inc., Pawtucket, RI, USA) using
dimensions from CT scans of impacted crania Most elements of the cranial geomety
were accounted for, although some regions were simpliﬁed due to the distance tom the
area of interest and complexity of meshing (Figure 5.1). The curvature of the parietal
bone was determined from CT scans and a 5-point spline was used to simulate the
measured curvature. The average bone thickness of the parietal bone was taken from
specimens tested in 4-point bending and was altered with age, but the modeled bone was
assumed to have a uniform thickness. The sutures were given a more pronounced width
than observed on the actual cranium since the material properties used in the model were
not directly obtained from sutures, but involved the bone on both sides of the sutures as
well. The primary landmarks in the basal region were included, but the speciﬁc
undulations were removed and replaced by a ﬂat plate with the same thickness as the rest
of the cranium. A region of the frontal bones was modeled, but the snout and orbits were
removed. The model consisted of symmety about the sagittal plane, but not about the
coronal or tansverse planes. Each bone and suture of the cranium was modeled

separately and constained at the bone-suture interfaces to form the full cranial model.

70

 

Contact area (varied)

Parietal Bone

      

# FE Parietal
Sutures Frontal
«7",— B (smooth line)
~ one
Occipital
Bone

Foramen

Magnum Base Orbit* CT Parietal

Figure 5.1. Simpliﬁed cranial model with overlays of modeled stuctures on true
stuctures. Coronal cut of CT used for thickness comparison. *Feature not modeled.
TFeature simpliﬁed. #Suture width altered for consistency with material property data
obtained previously. Dotted line represents location of frontal bone removal.

Finite Element (FE) analysis was utilized in an effort to produce and examine the
expected stesses and stains developed on the cranium and to compare these stesses
and stains to experimentally observed fracture patterns with respect to both age and
interface. The thickness of the cranium and the material properties of the bone and suture
were altered in the model to represent changes in age. Age related changes in the elastic
moduli of bone and suture were taken from experimental results of a previous study

(Baumer et al., in press), and average impact force and contact area data from another

study were used to simulate the difference between a rigid and a compliant interface

71

(Chapter 3) (see Table 1). Brieﬂy, the rigid interface was constained to produce smaller
contact areas than the compliant interface with both interfaces producing the same
magnitude of impact force. Numerous trials were run within the elastic range of bone to

develop a representative and clearly characterized pattern of principal stess‘and stain

 

 

directions.
Table 5.1. (Average) properties used in development of the FE model and impact
conditions.
Elastic Modulus Impact Force Contact Area
Specimen Age Esut Ebene Rigid Compliant Rigid Compliant
Young (3-7 days) 3.38 7.21 561 508 94 119
Old (18-21 days) 3.37 5.29 930 860 176 322

 

 

 

 

 

 

The magnitude of principal stress and stain directions were ﬁltered to remove
the lowest 20%. The remaining stesses and stains were compared to the experimentally
produced fracture patterns. A representative cranium from young and old specimens
impacted by each interface was selected. Each chosen cranium displayed all the common
fracture sites from the age/interface and no instances of uncommon sites. A photo
editing program (Photoshop ..) was used to overlay the modeled stesses and stains onto
a photograph of the chosen, experimentally ﬁactured cranium.

RESULTS

An overlay of maximum principal tensile stess directions revealed a
concentation at the impact site for all age-interface combinations (Figure 5.2). The
length of the arrow at each location represents the relative magnitude of stess at that
location. Comparison of these overlays revealed distinct lines of expected fracture
between suture boundaries and the impact point, regardless of age or interface. However,

the proximity of the stess ﬁeld to the boundary varied by interface. Speciﬁcally, larger

72

 

regions of high tensile stesses were located nearer the suture-bone boundary for
compliant interface impacts than for rigid interface impacts. Furthermore, tensile stesses
developed in the sutures of the young specimens were signiﬁcantly greater for compliant
interface impacts than for rigid interface impacts. The old specimens had similar levels

of stess in the sutures for both interfaces.

 

Figure 5.2. Overlay of maximum principal tensile stesses from FE models on
representative experimental cranial fractures for (a) rigid interface impact to a young
specimen, (b) rigid interface impact to an old specimen, (c) compliant interface impact to
a young specimen, and (d) compliant interface impact to an old specimen. Longer lines
represent higher stesses.

An overlay of the directions of maximum principal tensile stain revealed high
stains at or near bone-suture boundaries, with lower levels of stain at the impact site for
all age-interface combinations (Figure 5.3). The rigid interface impacts produced higher
levels of stain closer to the impact site than the compliant interface impacts, however,

the highest strains were still located away from the impact site. Furthermore, the

73

magnitude of the tensile strains developed in the sutures of the young specimens from
compliant interface impacts was considerably greater than those from rigid interface
impacts. The old specimens had similar levels of stain in the sutures for both interfaces.
Comparison of these overlays revealed some possible paths of expected fracture between

bone-suture boundaries and the impact point.

 

Figure 5.3. Overlay of maximum principal tensile stains from FE models on
representative experimental cranial fractures for (a) rigid interface impact to a young
specimen, (b) rigid interface impact to an old specimen, (c) compliant interface impact to
a young specimen, and (d) compliant interface impact to an old specimen. Longer lines
represent higher stains.
DISCUSSION

Modeling of a cranium with sutures resulted in a heterogeneous structure and,
therefore, regions of high stess did not necessarily correspond to regions of high stain.
Speciﬁcally, high stesses were generally located near the point of impact in the center

of the parietal bone, while high stains were located at or near bone-suture boundaries.

74

Experimental results showed that blunt force tauma to the parietal bone consistently
resulted in fracture at the bone-suture boundaries of the parietal bone (Chapter 3), which
correlated well with areas of high stain. The correlation of fracture to stain has been
previously documented in a study on failure of vertebral sections by Silva et al., (1998).
The study suggests that regions of high stain may be a better predictor of failure than
regions of high stess for heterogeneous stuctures. However, the current study utilizes
two distinct materials resulting in different failure criteria for each. It has been shown
that stains generated in the suture of the developing porcine skull during normal
frmction can be 6 times greater than the stain developed in the bone (Herring and Teng,
2007) and in 4-point bending failure tests the stain in the suture can be 2 to 3 times
greater than the stain in bone (Baumer et al., in press). Thus, it would be expected that
initial material failure would occur in bone prior to suture for conditions of similar
stain. Therefore, the stains in the bone near the boundaries of the parietal bone, which
were slightly lower than the stains developed in the sutures, likely resulted in failure
prior to failure of the suture stucture in most cases.

Age effects played an important role in the distibution of stains on the cranium.
The young specimens showed more frequent and greater magnitudes of stain in the
sutures than the old specimens, which correlated well with the experimentally observed
patterns of suture failure (Chapter 3). Speciﬁcally, the developing skull ﬁ'om birth to
sixteen days of age showed signiﬁcantly higher rates of suture failure than specimens
over sixteen days of age. The old specimens showed much higher ﬁ'equency and
magnitude of stain within the bone than the young specimens with some lower

magnitude stains in the sutures. This shifting of stain from the sutures to the bone with

75

increasing age was likely related to the changes in suture properties observed between
fourteen and eighteen days of age (Baumer et al., in press). Overall, the modeled patterns
of stain seemed to support the general characteristics of experimentally generated
ﬁacture with regard to age.

The compliant interface produced high stains within and near the sutures while
the rigid interface produced more localized stains around the impact site with lower
magnitudes of stain in the sutures. The larger contact area during compliant interface
impacts seemed to result in strains distributed nearer the parietal bone boundaries. This
difference in the proximity of stain in the bone to the bone-suture boundaries may help
explain the greater extent of fracture in compliant interface impacts at young ages.
Speciﬁcally, higher levels of stain near the boundaries may have promoted propagation
of fractures initiating at the boundaries, resulting in greater damage to the skull from
compliant interface impacts.

Despite the difference in proximity to the parietal bone boundaries, both
interfaces produced higher levels of stain away from the point of impact than at the
point of impact. At young ages, both interfaces produced high stains that were largely
constained to the coronal and temporal sutures. However, suture damage was only seen
in 29% of specimens aged under 16 days, while bone fracture occurred in 97% of these
specimens (Chapter 3). This suggests that the sutures were better able to absorb the high
stains while the slightly lower levels of stain observed in the parietal bone resulted in
material failure. Of course, some suture damage was generated experimentally,
suggesting that the sutures were not always able to fully absorb the high stains. It is

currently unclear if the sutures in these cases failed before of after the bone, but it may

76

be reasonable to assume, based on the modeled stains, that the suture damage occurred
after bone failure. The large stains developed in the suture model with relatively low
ﬁequency of experimental damage may suggest that prediction of suture failure on a
case-by-case basis may be difficult. However, the correlation between a reduction in
magnitude of stain and the disappearance of damage in the sutures for older skulls
suggests that some element of predictability may exist.

In summary, a simple model of a porcine cranium subjected to the effects of
experimentally induced blunt force tauma to the parietal bone was largely able to
predict characteristics of observed fracture patterns. The maximum principal stain
directions generated fnom the model seemed to correlate well with locations of ﬁactures
seen in a previous study (Chapter 3). The results of this study were not meant to propose
that the theoretical stesses or stains developed on this simpliﬁed cranial model actually
predict an arbitary fracture pattern developed experimentally. The results do show that a
simpliﬁed analysis of tensile stains and, to some extent, stesses developed ﬁ'om some
basic information on material properties of the cranium and interface speciﬁc loading
patterns can generate patterns that are consistent with some aspects of experimental
fractures. The experimentally obtained ﬁactures in conjunction with the FE model
suggest that blunt impact tauma to the parietal bone of the developing cranium leads to
predictable ﬁacture locations. This information may prove relevant in determining
perimortem events in forensic cases of head injuries to the pediatric patient with

unknown or questionable circumstances.

77

REFERENCES

Baumer T, Passalacqua N, Powell B, Fenton T, Haut R., Accepted for Publication, “Age
Dependent Mechanical Properties of the Infant Porcine Parietal Bone and a Correlation
to the Human,” Journal of Biomechanical Engineering, in press.

Billmire M. and Myers P., 1985, “Serious head injury in infants: accident or abuse?,”
The American Academy of Pediatrics, 75, pp. 340-2.

Case M, Graham M, Handy T, Jentzen J, Monteleone J ., 2001, “Position Paper on Fatal
Abusive Head Injuries in Infants and Young Children,” The American Journal of
Forensic Medicine and Pathology, 22, pp. 112-22.

Dickerson J. and Dobbing J ., 1967, “Prenatal and postnatal growth and development of
the cental nervous system of the pig,” Proceedings of the Royal Society of London,
166(1005), pp. 384-95.

Doorly M, Gilchrist M., 2006, “The use of accident reconstuction for the analysis of
taumatic brain injury due to head impacts arising from falls,” Computer Methods in
Biomechanics and Biomedical Engineering, 9(6), pp. 371-7.

Duhaime A-C, Eppley M, Margulies S, Heher K, Bartlett S., 1995, “Crush Injuries to the
Head in Children,” Neurosurgery, 37(3), pp. 401-7.

Frank F, Lawn B., 1967, “On the Theory of Hertzian Fracture,” Proceedings of the
Royal Society of London, 299(1458), pp. 291-306.

Graham R, Oberlander E, Stewart J, Grifﬁths D., 2000, “Validation and use of a ﬁnite
element model of C-2 for determination of stess and ﬁacture patterns of anterior
odontoid loads,” Journal of Neurosurgery (Spine l), 93, pp. 117-25.

Herring S. and Teng S., 2000, “Stain in the Braincase and Its Sutures During Function,”
American Journal of Physical Anthropology, 112, pp. $75-93.

Pierce M, Valdevit A, Anderson L, Inoue N, and Hauser D., 2000, “Biomechanical
Evaluation of Dual-Energy X-Ray Absorptiomety for Predicting Fracture Loads of the
Infant Femur for Injury Investigation: An In Vito Porcine Model,” Journal of
Orthopaedic Trauma, 14(8), pp. 571-6.

Prange M, Luck J, Dibb A, Van Ee C, Nightingale R, Myers 8., 2004, “Mechanical
Properties and Anthropomety of the Human Infant Head,” Stapp Car Crash Journal, 48,
pp. 279-99.

Prange M, Meaney D, Margulies S., 2000, “Deﬁning brain mechanical properties:
Effects of region, direction, and species,” Stapp Car Crash Journal, 44, pp. 205-13.

78

Prange M. and Margulies S., 2002, “Regional, Directional, and Age-Dependant
Properties of the Brain Undergoing Large Deformations,” Journal of Biomechanical
Engineering, 124(2), pp. 244-52.

Schneider L, Lehman R, Pﬂug M, Owings C., 1986, “Size and Shape of the Head and
Neck From Birth to Four Years,” Washington, DC. The Consumer Product Safety
Commission, Report No.: UMTRI-86-2.

Silva M, Keaveny T, Hayes W., 1998, “Computed T omography-Based Finite Element
Analysis Predicts Failure Loads and Fracture Patterns for Vertebral Sections,” Journal of
Orthopaedic Research, 16, pp. 300-8.

Thibault K. and Margulies S., 1998, “Age-Dependant Material Properties of the Porcine

Cerebrum: Effect on Pediatic Head Injury Criteria,” Journal of Biomechanics, 31(12),
pp. 11 19-26.

79

CHAPTER SIX

CONCLUSIONS AND RECOMMENDATIONS FOR FUTURE WORK

The previous chapters describe the material and fracture characteristics of the
developing porcine skull. The bone and suture properties were obtained through 4-point
bending experiments on beam specimens removed from the cranium. The parietal bone
was impacted under contolled laboratory settings to produce ﬁ'acture patterns with

regard to age and surface interface. A simpliﬁed model of the infant cranium was

 

developed in a ﬁnite element program for comparison of theoretical stesses and stains
with observed fracture patterns. The model was used to simulate clinical cases of
bilateral crushing injuries as well as blunt impact tauma to the parietal bone.

In Chapter 2, beam specimens cut tom porcine crania were tested in 4-point
bending at 25mm/sec until failure to obtain material properties of the bone and suture.
While the mechanical properties of parietal bone and coronal suture did not change
signiﬁcantly with age, bone specimens showed an increase in bending stiffness with age.
Bone-suture-bone specimens showed an increase in bending stiffness only between 14
and 18 days of age. Correlation analyses using existing and new data to compute the
bending rigidity of infant parietal bone specimens suggested that days of pig age may
correlate with months of human age during the most common time ﬁarne of childhood
abuse cases. Future studies should investigate the possible rate sensitivity of the
developing bone and suture, speciﬁcally focusing on high rate loading to simulate
impact rates.

In Chapter 3, skull fracture on infant porcine specimens with respect to age and

interface were documented in an effort to identify fracture characteristics for a

80

 

developing skull under known conditions. Fractures on the infant porcine skull due to
impact with a ﬂat surface initiated at points outside the zone of contact for both rigid and
compliant impact interfaces. The skulls of young infants suffered relatively more
extensive tacture damage stiking a compliant versus a rigid interface for the same
impact energy. As the animal aged, however, the damage caused by the compliant
interface decreased until no damage was present at energies causing fracture with a rigid
impact interface. Future studies should examine additional interfaces of impact,
especially involving common surfaces of potential accidental injury such as carpeting or
playground rubber. Additionally, the extent of propagation and fracture patterns
developed under higher energy impacts should be documented.

In Chapter 4, four clinical cases of fatal crush injuries to young children between
1.5 and 6 years of age were documented with correlations between modeled stess and
clinically observed ﬁacture patterns. Regions of high tensile stess in all four cases were
located on the basicranium and appeared to follow the ﬁacture path(s) extending tom
the location of maximum stess. These results suggest that pre—failure stess ﬁeld
diagrams may give some insight into the projected fracture propagation paths, although
prediction will not be exact. Also, these analyses of the modeled stesses and the
clinically observed ﬁactures suggest that quasi-static bilateral loading of the cranium
may lead to predictable fracture of the basicranium, possibly independent of age.
Additional case studies would be helpful in validating the model further. Future studies
may also be able to obtain more accurate results using CT scans of infant crania. With
CT scans, the actual density distribution within the skull could be accounted for rather

than the approximation used in this study. Also, the skull dimensions and basal landmark

81

 

geomety were approximated. Approximate thickness changes with age were the only
geometric property altered.

In Chapter 5, the impact conditions generated experimentally for rigid and
compliant interface impacts to porcine skulls of various ages were recreated on a
simpliﬁed ﬁnite element model of a cranium. Experimental results showed that blunt
force trauma to the parietal bone consistently resulted in fracture at the bone-suture
boundaries of the parietal bone, which correlated well with areas of high stain. The
theoretical stains also showed similarities with the signiﬁcantly higher rates of suture
failure in young specimens than in old specimens. Interestingly, the larger contact area
generated during compliant interface impacts also seemed to produce stains within the
bone distributed nearer the parietal bone boundaries, which may have promoted
propagation of fractures accounting for the greater amount of observed damage on the
cranium produced from compliant interface impacts. Future studies should develop a
model of the skull from exact dimensions of a CT scan.

The information presented in this thesis documents baseline information for
ﬁ'actures generated on the developing porcine skull. An age-based correlation with
porcine specimens may provide an important biomechanical tool for the study of
pediatric skull fracture in efforts to investigate cases of childhood abuse. This
information may prove relevant in clarifying unexplained perimortem events in cases of
injuries to the pediatric cranium. Modeling cranial ﬁacture is particularly important in
unwitnessed injury cases or in cases where information tom examination of skin and
scalp is not possible due to skeletonization. These data present several factors that must
be examined on a case-by-case basis and may ultimately lead to the deciding factor in a

courtoom verdict.

82

APPENDIX A
RAW DATA FROM CHAPTER 2

83

 

 

 

00... 0 ...0 ..000.0 00.00.. .000 0 ...000 .00 P0. ..0 3. 0000
00. .. 00.0 0000.0 00.0? 000.0 v0.09. 00.0 00.0 3. ..000
00. F 0.0 00.0.0 00.00 .. 00 .. .0 00. ..00 0 0.0.. 00.00 3. ....00
00. .. 00. _. 3.00.0 00. .0? 00. ...0 .000 0 .. ....0 0.. 0000
0 ...0 0v. .. 030.0 00.00 F 000 .0 #0000 00.0.. 00.00 0.. 0000
00.0 00. 0 0000.0 00.3. 000.0 .0003. 00 ... 0.. .0 .1. .0000
#0.. .....F 030.0 00.00 0.120 00.....0 00.0.. 00.... 0.. 0.00
..0... 00.0 300.0 8.3.. 00 F .0 00.000 00.0.. 00... .. 0.. 0 ..00
0... .. 006 .0000 00.00 .. 80.0 00.000 00. .. .. 00 .0 0 0000
00... v0... 00.0.0 0 ....V0.. .00 .0 00.00. 00.0.. 00 ..V r 0 0.00
00... 00... 0000.0 00.00v 00.? 00.000 00.0.. .40.; 0 0.00
.V ..... 006 0000.0 0 ...00.. 00 .. .0 0v.000 0 w.0.. 00 .3. 0 0000
0... .. 00.0 0¢00.0 V0000 V0000 V0. .. _. 00.00 0 0000
00. .. 00.0 0.00.0 00.8 0 ..0.0 00.0.... 0 ...0.. 0 .. .0 .. 0 0000
00.0 3.0 00 ..0.0 00.00.. 8 .. .0 0.0000 00... v 00.0.. 0 0000
00. .. ..0.0 0000.0 0.0.000 00.. .0 0.0000 0 ...0.. .00.. 0 0000
0.0.? 0 ...0 .030 0 ...000 000.0 00.0 ..0 .00.. 0.000 0 0000
00... 00.0 0000 .0 0 ...000 8V0 00.00.. ..0..... 00.0.. V 300
V ...P 0.00 0.00.0 00.00.. 00 .. .0 0 0.000 00... .. 00.0.. v 0000
00.0 00... 0.00 .0 00.00.. 00 .. .0 00.000 00... .. .. ...0.. v 0000
00.0 0.0.00 0000.0 3.00... 000. .. 00.000 00 .00 00 .v0 0 @000
..0.0 0000.0 00. v0.0 00 .. .0 00.000 00.0.. 0.0.3. v 0000
.88. $00. €8.20 30.2. .3 .82. .EE.

$9.on 8332 868m 59m as 0.85 ...: Foam 58m 38505 328985

.2. .08.. a? 85.8%

 

 

.3088on 3000508050 me new»: 050:2. genre :80 028:8 8% 33. ...< 03:.

84

 

 

 

 

 

 

 

 

 

 

85

 

«3 SN Named 3.8 82. ~38 3m «5. a mean.
3..- 88 88a 8R mood ~38 8.4 :3 N Eda
9N Ea 888 8.0a Sod were 8.... E. a mean.
2a 3.. 288 8.2 mood 88a ta LB a mean.
8a and 823 8.8 Sod 8.8a 8s «3 a. amen.
84. 8a $55 8.8 Sod 82$ 5... 5m 2 Ram
New as 885 8.8 :3 9.52 :6 8s t 5%.
mod an. 888 «tea as! 3.8.. 5.9 ES 8 8a....
01 «3 83¢ 3.8..” $3 meta and. 8.9 a. 2am
Es. .aaov Eases. 802. 365$ .52. .55.
85.95 8.38.2 Pecan Ease ...: macaw ...: 52w «.8an 558935 2:83 a? 858%

 

 

 

 

 

 

 

 

 

 

cos—5:00 .~.< «Bah

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

8.: 3.0 8I~00 :88: 8% 00.08: 8.0 8.0: 8: 8%.
0:: 5.0 8:00 :50 80.0 8:8: 80... 0:8 8: :0%.
80.: 8.0 80.0 888 8.8: $8 8.0: : :%.
00.: 8.8 8:0.0 :8: 0.60 0%: 8.0 00.0 8 0:%.
8.: 8.0 080.0 8.8: 80.0 80.80 8.0 8.: 8 8%.
8.8 :8 080.0 8.8 80.0 8.08: :00 8.0 0: 00a...
00.: :0 080.0 8.8 :30 ::.:::: 8.0: 8.8 0: 8%.
0:.: 8.0: @800 8.88 08.0 8.80 8.: 8.0: 0: 8%.
8.: 8.: :80.0 8.8: 80.0 8:0 00.: N08 0: 8%.
:N: 8.0 :::0.0 80.88 80.0 8.88 00.8 8.0 0: 3%.
8.: 88 080.0 8:8 80.0 8.80 8.: 80.8: 0: 8%.
8.: 8.0 0800 8:0: 08.0 8.08 8.0: 0:8 0: 8n...
8.: «0.0 00:00 8.8:. 83 8E 8.0: :08 0 8%.
8.: 8.8 8:00 8.80 :00 8.088 8.8 8.8 0 8n...
8.: «0.0 8:00 8.88 80.0 8.80 8.: 8.0 0 8%.
8.: :08 88.0 8.08 8:0 8.08 8.8 00:0 8 00%.
8.: 8.0 80.0 8.80 80.0 8.88 00.: 8.8 8 Sn...
8.: 8.8 080.0 8.80 80.0 8.08 88 00.: 0 00%.
8.: M08 0800 8.5 :00 8.88 8.0 8.8 0 00%.
8.0 8.: 880.0 8.800 80.0 :38 8.0: 8.8: 0 8%.
8.0 8.0: 880.0 8.88 08.0 00.08 8.8 8.8 0 8%.
00.: 8.0 080.0 8.8 08.0 8:8 8.0 00.8 8 8%.
8.0 :8: 88.0 8.88 80.0 8.800 8.0 8.0 0 88
8.0 :E 88.0 8.80 80.0 8.08 00.0 8.: 0 8%.
80.: 8.0 800.0 8.08 ::0.0 8.08 8.0 8.8 :. 8%.
00.: 8.0 00:00 00.0I0w. 80.0 00.80 8.8 8.0 v 8%.
AEEV Ann—0v AEE\EEV 3&2 g. 36cm AEZV AEEV
m8§2£ 8.38.2 8.0000 0.90m .05 880.5 0.80 w8080 000580.85 2.08.. 03 005.080

 

 

858608 ocoo. 8339833 MO $08 9:083 «509+ :88 380:8 33 30m .N.< 239

86

 

 

 

 

 

 

 

 

 

 

 

87

 

 

 

 

 

 

 

 

 

 

 

8d mg 955 2.3 «8.0 8.29 NS. 3d a man.
one «:3 8.8 Bod 52$ 8m «mm a 3%.
Ba 8+ 885 3% mad 2.39 96 8d a 9%.
and mom 823 8.8 $8.0 «3 8.2 E «Eu
3.. an 835 2.5 «36 ~33 3.» m3 2 8%.
m? Re 25o 8.2. Bod «tamﬁ m3 3m 2 8%.
one 385 2.0: mood $.88 8.» 5o 2 NE
SN 8s 885 8.5? Bod 3.” mod S 3%.
84 So 285 8.8 Sod $.on an EN 2 2%
a: 3m E86 :aﬁ «Mug 38$ 8.» 3.2 3 Ram
AEEﬁ ANQOV $55595 awn-2v A3 5cm AEEV AEEV
”85.2.: 3:32 96:8 526 .5 33w .5 595 amoéam “528335 2 83 o? cogoogm

 

62:5ch .~.< 033,—.

 

 

 

..0.0 00.0 0000.0 8.00 000.0 00.0 0v.0.. 3. 3mm
00... 3.0 00 50 00.000 00.0.. 00.0.. 3. 001m
00. .. ..0.0 800.0 006 ..0 00.005 00.0.. 00.00 S. ..0n.n_
..h. .. h ...v 00 ..00 00.00 500.0 00 .000 00.0 00..q 3 ....Qn.
0V. .. 00K #0 ..00 00.000 000.0 00.000 00.0 ..0.0 0.. 0..n.n.
00. .. 00.0 ..n ..0.0 00.00.. ..000 $.03. ..v.0 00K 0 .. 00n.n.
0n. .. 00K 00.00 000.0 00.00 .... _. ...0.. 5.00 0.. 00am
00.0 000 00 ..00 0 ...00.. 000.0 00,000.. 0.}‘ 00.0 0.. 00n.n_
00. .. 00... 00 ..00 00. v0? 0000 .000.» 00.0 010 0.. 0_.n_a
0 ... .. RN ~“.0000 00.00? 00.0.. 0.000 or 0..n.n.
00. .. 00.0 00 ..00 3.000 000.0 00 ..K0 .0... .. 00.0 0.. Sin.
00... 00.0 0.. ..0.0 00.00 000.0 0v.000 0h .0 510 0.. 0am
00. .. 00.0 0000.0 000 ..0 by .000 0 .. .0 .. 00.0.. 0.. 0......n.
00. .. 00..‘ 0h ..00 00.03 5 ...0 00.0 ..0 00.0.. 00.0 0 0_.n_n.
0..... 0.00.. 00 5.0 3.. ..0.. 05.0 50.30 50.0 50.0 0 tan.
#0. .. 00.0 ..0 ..00 00.00 V ..00 00.000 «0.0 50.0 h 00n_n.
..0.0 0000.0 010.0 0 ..00 3000 00.5 00.0 n noun.
00. .. 0.00 #0000 00.00.. 000.0 0.000.. ..00 V0.2. 0 00mm
00.0 00... ..0000 000 ..v 0 ..00 00.0 0 0.20.. 00.0? 0 00n.n.
00.0 00.0 #0 ..00 00 ...0 00.90 0.0.. @0.3 0 vomn.
N0. .. 00.0 0000.0 0000.. ...V00 0.. .000 00.0.. 00.0.. 0 81¢
3V. .. 00.0 00 ..00 00.00.. 0 5.0 2.000 3 N 00.0 v 00n.n.
00.0 00.0 0000.0 00.0 ..0 000.0 00. .....0 00.0 {.0 0 00mm
8.0 00.0 00 ..00 00.000 0.000 0 .. .000 010 .. 00.0 0 01¢
00. _. 00.0.. 0000.0 00.000 v 5.0 5.000 00.0 N00 v 00.:
.0; 0 ..K 00 ..00 v ...000 v ..00 0.000% 00.0 0 V0 v 00%.
.EE. ammo. €5.88. Ana—E A... .82W .EE.

822%: 3.822 86:8 595 .5 83m ...: 3.8m 59% 88:5 23883

.20 83 a? 553m

 

 

£88.30». 280 .2330 ..o 38. 050:2. 3.00-.“ SE... 098:8 8% 33. .03.. 0.05..

88

 

 

 

 

 

 

 

 

 

 

 

o..~ 8.” 8.3. 8.... .3... 3.08. .3 3.. .N 3%.
...... on... $5... 8.8. as... $33. mom was .N 3a....
a... 8.. .32. «I: Sod .35 8... o... .N my"...
mum 2a 2.2. 2.8 .5... 3.8.. 93 8.. .N NE:
m... an... .~.o.o 8.3 Bod 8.8.. 8... ......” m. 8a....
6.. 8.. ...od N33 .8... «33. men an m. an:
3.. R... 38... 8.8 m8... 8.3.. and B... t .Nna
8.. ...m 0.8... .....m ...od 8.8.. om.» :3 m. 8%
8.. 8d 88... 8% mm... 3.8.. 8... SN w. 2%.
AEE. 3&0. AEEEE... 3&2. 0... 5cm $52. ASE.
«.852... 3332 9.68m swam ...: 32% ...: 52w 8&5 2958.35 .2. 83 on... 8588

 

 

 

 

 

 

 

 

 

 

 

02:55.0 .0.< 0..—ah.

89

 

 

 

3‘. F 00.0 08.0 00.05 00.0 «N000 v06 3.0 N 335
05—. _._..N #80 00.000 00.0 00.8w 00.9 0.. RN 0.. noun—n.
00... 00.N 000.0 v0.Nm_. 0.. .0 00.8» 00.2. 00.0w 0_. gown—n.
05.? 50... 08.0 0 _. .000 00.9 8.50m SN? 3200 0.. 0 En.
00... 00.N m 5.0 «5505 0.. .0 9.50m 50.? _. 00.3. 0.. v En.
0v; NNN 50.0 05.0NN 0N.0 «0.8N N _. .0_. 0.. .0? 0.. N En.

00d 8N0.0 50.000 «5.0 NEEN 00.N_. 9.0—. 0.. Sin.
00. _. 00... 3.00.0 mm. .0? 30.0 00. SN KN? m —. .vw 0? 00am
5N. _. mod h 5.0 50.0NOF v0.0 03.0mm 00... _. ..0.v~ 0 00am
00.0 ..NN 0 v0.0 3.0 ..0 5.0 0N.00 MK .0 _. No.0? 0 Von—n.
v0.0 0 ...0 v v0.0 50.000? 00.0 «0.2.0 NONF 5.0m V sown—n.
00.0 00.0 _. V0.0 N0 .000 :20 00.0VN F0... _. 5.0—. 0 mun—n.
A880 089 AEEEEV ﬁns: 3 0520 AEEV

3.8on 3382 965m 226 .5 32% ...: 3.8m gnaw chEm 52.8335

20 28.. o? 559%

 

 

.825 833. E mat—smog 0.02 2.258% oconéhﬁambcoa .00 3m»: 0505: “500% 890 038:8 83 Bad 6.4 03:.

90

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

91

8a mg 83 8.8 5.0 8.08 80 N3 3 38%.
m3 F3. 23 9.8 55 8.8: 23. 8.5 8 39.;
3.: 3.0 :3 8.3 So $.32 23 5m 3 £3;
SN 8.” 22. 8.8 36 8.2a .3. m3 2 £8“.
«2 8.0 83 ”EN 86 8.89 «B 8.: 3 88¢
o? 53 8.: 5o {.33 5e 20 3 32“.
m: E 93.0 8th 5.? $.58 E: 3.8 3 8.8...
NE SN Rod 83 Q3 9.3” F; 8a 3 «NE;
8.? Ea BS :3” ”3 EN: :2 8% 3 £8;
m: Bu ~86 SN... 85 3.8,. 8.» 8.0 3 «5%.
:4. ta 82. 8.8 So 3.8m 8.5 8.0 2 88.”.
a: 955 No.8? 3.0 33% 8.2 3.8. 2 «3%.
m: «.3 $3. 3.8? 2.0 2.9% 3.: 8? 0. 88a
«2 8.” 83 $.52 ”We 5.8.... 3.9 8.0 .V «8%.
2:5 ﬁne 3.52.5 was 3 8m E20 :55
$05.05 3.332 320m 595 .5 $26 .5 :05 $9an $58520 A20 28.. m9. cmgoam

 

 

 

 

 

 

 

 

 

 

 

 

.8355 25$ 20 E00 :33 25863.9. econ—653-080 20038 .80 $30 050:3 0500-0 80.0 028:8 300 Bad .m.< 0:35

Table A.6. Porosity data from stained cross-sections of perpendicular bone specimens.

 

 

 

Bone Total Porosity
(Pixels) (Pixels) (%)

PP02 33547 41699 19.5%
PPO3 25797 31059 16.9%
PP04 61503 68346 10.0%
PP05 69804 81977 14.8%
PP06 76844 79827 3.7%

PP07 52030 56810 8.4%

PP08 45056 50665 11.1%
PP09 43168 56322 23.4%
PP10 43379 62302 30.4%
PP12 37707 43774 13.9%
PP13 67578 74024 8.7%

PP14 66050 76634 13.8%
PP15 29341 34137 14.0%
PP16 66006 85252 22.6%
PP17 29099 33402 12.9%
PP18 87797 100901 13.0%
PP19 75149 99736 24.7%
PP20 67324 86982 22.6%
PP21 84893 105391 19.4%
PP22 76138 84575 10.0%
PP23 52362 57918 9.6%

PP24 10420 1 1466 9.1%

PP25 41367 45319 8.7%

PP26 36817 40503 9.1%

PP27 91413 111205 17.8%
PP28 55797 71386 21.8%
PP29 62829 72158 12.9%
PP31 109448 1 18586 7.7%

PP32 123319 160033 22.9%
PP33 55829 61288 8.9%

PP35 111136 139891 20.6%
PP39 236070 305571 22.7%

 

 

92

Table A.7. Porosity data from stained cross-sections of parallel bone specimens.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Bone Total Porosity
(Pixels) (Pixels) (%)

PP02 59955 74687 19.7%
PP04 23619 25450 7.2%

PP05 32821 38448 14.6%
PP06 121544 136713 11.1%
PP07 47076 51903 9.3%

PP08 72309 85020 15.0%
PP09 52524 58510 10.2%
PP10 69059 86167 19.9%
PP11 135612 168833 19.7%
PP12 68397 76470 10.6%
PP13 67312 76059 11.5%
PP14 62774 72731 13.7%
PP15 67010 77393 13.4%
PP16 50708 66943 24.3%
PP17 65382 71443 8.5%

PP18 89707 97636 8.1%

PP19 51343 65791 22.0%
PP20 1 12397 137864 18.5%
PP21 39143 47067 16.8%
PP22 51157 57322 10.8%
PP23 27210 29493 7.7%

PP24 12117 13427 9.8%

PP25 21 753 24577 1 1.5%
PP27 67091 105630 36.5%
PP28 37794 42043 10.1 %
PP29 72273 85481 15.5%
PP30 51137 56511 9.5%

PP32 100422 125993 20.3%
PP34 122032 141013 13.5%
PP35 146688 168171 12.8%
PP39 5921 19 889588 33.4%

 

 

93

APPENDIX B
RAW DATA FROM CHAPTER 3

94

 

mm

 

 

8.8 8.. 8.8~ .8.” 8888 8.8 8.88 88 V. BE
8.8. K. ~88 .88 880.0 85. 8.88 88 8 ... 08E
8- 8.. 8.88 o8.» 88... 8.8 2.88 88 8 ~. 88:
m... 8.. 8.8. ~8.~ 88... 8.8 8.88 88. 8 ~. 82..
88 8.. ~..8. ~8.~ 88.0 ~..o. ...8~ m8. 8 .. 82¢
-- 8.. .38 88.. ~83. 8.8 8.8.. 88 8 8 82m
8.8 8.. 8.8 8~.~ 88.0 8.» ~38 88 8 8 ~85
8.8 8.. 8.5.. §.~ 2.8.0 8.8 8.8.“. 88 8 m .8:
8.8 ~... 8.8. ~8.~ 88... 8.8 8.88 88 8 h 88:
8.8 .... 8.8 ~8.~ 888 8.8 8.8 88 8 8 ~85
...8 8.. 8.8 ~8.~ 88... 88 8.88 88 8 8 82¢
.8. 8.. 8.8. .8.~ 88... 8.8 8.88 88 8 o 82..
o.-. .... 2.8. 8¢~ 888 o... 8.88 88 8 m 82..
8.8. 8.. m~.o.~ ~88 .83. 8.8 8.8 88 8 8 ~88.
8.8. 8.. 8.8. ~8.~ 888.0 8.8 8.5, 88 8 m 82...
8.28 8.. 8.8. 8.8 880.0 8.8 8.88 88. 8 .. 82a
8.8 8.. 8.8. 8- ~80... 8.8 888 88. 8 m 82“.
Q8 3.. 8.8 ~8.~ E88 8... £88 88 8 m .8...
8.8. 8.. 8.8 8.. 8808 8.0 8.8.. .8. 8 ~ 08:
8.8 8.. 8.8. 8.. 88.0 E... 8.8.. v8. 8 . ~ 88:
-- -.. 8.8.8 88... .83. 8..., 883 .8. 8 ~ .8:
£55 8.5 8525 3.82m E 88.8 8.5 83 ea: .2. 83”. a. 8a: 85 8.89 85.883
82 $8on 8858 2.8“. .085 8 52.8885 .08.... 8.8“. .88: 85 8<

3862

 

 

.88.... 8.83:. a»... 828 8.02.8 38 as. ...m 2.3.

95

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

odov N. ..m modmm .V. ..N _ m.oo.o _ an...» 8.33. mo .N mm mm m. .5
BEN end .0. .8 wood owood 8.0 Eden. oo.w ow mm x. . ._n_
QR mm. not . w mEmu =38 - 8882 .0: Emu .cOzocémE .oSano moo. mm m. omoE
8.8 8.. 8:89 085.8 =33 - 8282 .9. 98 .5888... .28 .80 88 8 ~ 0 SE
«.mm 86 notomm. wmemu =83 - 3882 no: 8% 60:05.8... .55an0 mom. on H oon
«.5. RN 3.8. m3... 886 .mN. Node... m5. om mm on ._n_
3.3 n ..m 8.0.... mmmd ovood .wd. wndmm. m5. om mm cm :1
mdmm vmd NmNo. $5.0. «mood 3N. wvdmw. m .m. on om mv ._n_
m. ...N 3N 3.3. Nam...“ 9.8.0 o... .. 5.3.0. m5. om mm 8 ._n_
mdo. mm... 5.09. m .m.m Nmood ts 868 m5. on em ov ._n_
nee 3.. 8.3% mom. m .86 pm... 8.08 m .m. 0.. wm an ._n_
Qmmm Q... m ..EN 8.6 38.0 .vd m5. on «N 3:;
5.9. 3... Q... .m. EON 58.0 and 8.3K m5. om mm 9. .E
Qo§ $8. $58 ~86 good mmd. ondww m .m. om om 95?.
9%. 3.. 8.8. 30... omood and 0.95 80. mv m. N. ._n_
3%. 3... 868 mom. .Nood mo.m ...08 wow. mm 5. 59¢
98. 8.. mm. .m. End Rood and 3&5 80. mm o. w .oE
@va 8.. NNNE 08.. owood 09m 8. . .m 80. mm m. 391
me. 2.. modmm $68 886 m ..m modmw moo. mm m. amen.
A~EEV .EEV €28 3 8.8m 3 cozlmso Ass. 83 xm.‘ A2. 85“. 8. 84.5. Es ENE :2 _ Spam
8?. 30.2th 30:85 228“. .08.; um EwEoomﬁwE .08.; 9:8”. 290: no.0 09..
mmmhm><

 

 

 

 

 

 

 

 

 

 

 

 

62.5250 ...m Saab

96

Table B.2. Skull damage measurements for rigid interface impacts.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Bone Suture Total
Fracture Damage Damage
Specimen (mm) (mm) (mm)

P001 8 0 8
P1003 11 0 11
P1050 31 0 31
P1004 47 0 47
P1079 47 0 47
P1016 4 0 4
P1023 38 0 38
P1062 7 0 7
P1076 0 0 0
P1013 28 0 28
P1049 26 22 48
P1007 25 0 25
P1008 25 0 25
P1071 80 0 80
P1032 17 0 17
P1066 8 12 20
P1057 66 10 76
P1030 25 35 60
P1045 15 21 36
P1046 21 30 51
P1077 88 0 88
P1028 20 0 20
P1064 79 10 89
P1018 22 0 22
P1067 83 10 93
P1112 5 0 5
P1053 29 0 29
P1142 10 0 10
P1144 10 0 10
P1138 60 0 60
P1140 85 0 85
P1146 105 0 105
P1148 85 0 85
P1134 0 0 0
P1136 85 0 85
P1080 30 0 30
P1010 14 0 14
P1020 23 0 23

 

97

 

 

 

 

 

0.00: 2.0 8.80 :80 3.8.0 8.0 00.08: 82 00 2 80.0
0.08 8.: 8.000 000.0 008.0 8.: :0.:0: 82 8 0: 20.0
00:0 8.: ::.0:.: 80.0 88.0 8.:: 8.80 82 8 0: 20.0
0:8 :0: 000:: 8:0 880 ::.:: 00.000 80: 00 0: :8:
0.0 :0 8.: 000:: 20.0 88.0 00.:: 80:0 82 8 N: :00:
:8: 8.: 8.0:: 0::.0 88.0 2.:: 8.08 82 8 0: :00:
:0: :0: 8.:0: 08.0 88.0 8.: 8.80 80: 00 :: 80E
«:0 8.: 8.:0 0:00 880 :02 8.08 82 8 2 000.0
0.80 2.: 00:0: 08.0 88.0 8.0 00.8: 82 0: 0 08.0
0.000 8.: 00:2 80.: 008.0 8.0 000:0 82 00 0 «:0:
0.8: :00 8.00 80.0 :080 8.: 8.20 82 0: : 08.0
0.00 :0: 0:0: 0::.: 008.0 00.0: :88 82 0: : :2“.
0.8 8.: 8.8 80.: :080 8.0 2.80 82 00 0 08.0
0.0: 8.: 8.0: 0:00 0:80 8.0: 00.80 82 00 0 0:0:
0:0: 000 :08 02.0 88.0 8.: 8:8 82 00 0 08.0
:8: 8.: 8.:: 80.0 :080 8.0 808 82 00 0 80:
0.00: 8.0 :08 80.0 0:8.0 :0 8.8:. 82 00 0 08.0
0.00 8.: :08 0:00 88.0 8.0 8.80 82 00 0 0:2...
0.8 00.0 8.0: 80: 008.0 8.0 00.08 82 0: 0 08.0
0.08 00.: 8.0: :00: :80 8.: 8.08 82 00 0 :00:
0.00 00.: 2:0 0::.0 88.0 :00 8.00:. :8: 0m 0 08.0
0.0: 2.: :08: 0::.: :080 ::.0 00.20 80: 0: 0 80.0
:80 00.: 88: 0:0: 88.0 0:.0 0:.:0:. 82 0: N 08.0
0.8: :0: :30 08.: 88.0 8.0 2.08 80: 2 0 80.0
meg 8.5 828 3 3080 888050 8.5003 x02 A2820“. av 0002 0:9 8:03 85800
8a 085.2: 08080 05:0". 8005. :0 8580.85 .805. 05:0“. 203: 020 00<

omEo><

 

 

.3805: 8835 0800: 82 a: 080088 :50 02808 300 300 .0:— 20.;

98

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

00: ::.0 v0.00: 000.0 300.0 00. :: :0.:00: 00:0 00 00 00 :_n_
0. :0:. ::.0 :0.:.0 :000 0:00.0 0:. :0 00.000: 0 :0: 00 mm :0 :E
0.0:: : 00.0 v :.0: 0 ::.0: 0000.0 ::.00 :0. :0:: 0 :0: 00 00 00 :0.
00:0 0:.0 00.3: 00:.0: 0000.0 00.:: 00.00:: 0 :0: 00 00 03.1
0.00: 0:. : :0. ::: :000 030.0 :0. :: 0:000 0 :0: 00 00 :v :n.
0.:0v: :0.0 00.00: 00:.0 0000.0 :00: 00.:00 0:0: 00 v0 :35
0.000: 00.: 000.00 000.0 0000.0 ::.0: 00.00: 0 :0 : 0:. v0 00 :E
0.::: 0:.0 v0.00: 00:.0 300.0 00.0 0:00: 0 :0: 00 00 0:. :5
0:0:. 00. : 00. : :: :00.: 030.0 00.: 0:0: 00 00 00:5
0.0 :0 :0. : 00.00: 00 :.v 3000 00.0: 00.000 0 :0: 00 00 «00?.
0.00: :0. : :0.00: v ::.0 0000.0 00.0 00.0 :0 000: 00 :: 000E
0.0:0 00.: 00.000 :0:.0 0:000 00.0 00.00 000: 00 0: 000E
0.: :0 0:. : 00.00: 0.00.: 0000.0 0 :0 0:.:: 000: 00 0: 000E
:. :00 00.: 00.00: 000.: 0000.0 :00 00.:00 000: 00 0: 000E
:000 0:.0 00.00: 000.0 3.000 00.0 03000: 000: 00 0: :00_n_
€55 055 0520 0.10.000 E 000050 025 08.. x05. :20 020“. :00 80.2 0:3 3:03 002800

00.:: 002222.: 03:50 83:0”. 8008. :0 Em E80305 800 E. 05:00”. £081 no.5 00<

008020.

 

60:50:00 .05 030,—.

 

Table B.4. Skull damage measurements for compliant interface impacts.

 

 

 

Bone Soure Total
Fracture Damage Damage
Specimen (mu!) (mm) (min)

P1034 53 0 53
P1035 117 0 117
P1056 52 0 52
P1005 112 0 112
P1081 30 0 30
P1060 9 11 20
P1078 30 0 30
P1022 57 30 87
P1024 17 20 37
P1063 36 16 52
P1012 7 14 21
P1025 34 40 74
P1011 190 0 190
P1040 78 0 78
P1072 16 0 16
P1069 11 0 11
P1065 132 0 132
P1059 129 15 144
P1031 141 0 141
P1037 0 41 41
P1021 75 0 75
P1014 103 0 103
P1015 65 22 87
P1041 12 53 65
P1027 105 0 105
P1029 95 0 95
P1052 41 0 41
P1082 50 0 50
P1068 27 10 37
P1054 23 0 23
P1143 0 0 0
P1145 0 0 0
P1139 60 0 60
P1141 0 0 0
P1147 0 0 0
P1149 40 0 40
P1135 0 0 0
P1137 0 0 0
P1122 0 0 0

 

 

 

100

  
  

   

   

T

 

1| “11111111111111Wiles
0 3 0 6 2 L 8 3 2

3

 

 

..."..l..|.w."..
3 12 9