 

 

/1.

’ '7') 3' <3

/
f

[H533

This is to certify that the
thesis entitled

FRACTURE PATTERNS OF THE DEVELOPING SKULL
A'ITRIBUTABLE TO DIFFERENT IMPACT SCENARIOS

presented by

BRIAN J. POWELL

has been accepted towards fulﬁllment
of the requirements for the

MS. degree in EngineerinLMechanics

 

 

 

@Aﬂajor Professor’s Signature
W 0/ 0

Date

 

MSU is an Afﬁnnative Action/Equal Opportunity Employer

 

LIBRARY
Michigan State
University

 

 

 

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

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 K:/Prclecc&Pres/ClRC/Dale0ue indd

 

FRACTURE PATTERNS OF THE DEVELOPING SKULL
ATTRIBUTABLE TO DIFFERENT IMPACT SCENARIOS

By

Brian J. Powell

A THESIS

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

MASTER OF SCIENCE
Engineering Mechanics

2010

ABSTRACT

FRACTURE PATTERNS OF THE DEVELOPING SKULL
ATTRIBUTABLE TO DIFFERENT IMPACT SCENARIOS

By
Brian J. Powell

Forensic anthropologists and pathologists frequently rely on fracture pattern
analysis to determine the causation of trauma in pediatric abuse cases. However, due to a
lack of pediatric skull fracture tolerance data it is often difﬁcult to diagnose whether the
injury was inﬂicted or accidental. The research presented in this thesis utilizes an in situ
porcine head model and ﬁnite element analysis to assess the degree and pattern of skull
fracture in two different impact scenarios: entrapped and ﬁee fall. Chapter 2 documents
the skull fracture patterns generated due to a high energy blunt impact using a rigid and
compliant interface. The porcine specimens in Chapter 2 were entrapped in a bed of air-
hardened epoxy in order to prevent translation of the head during impact. In Chapter 3,
the porcine model was used to assess the degree and pattern of fracture due to a free
falling head impact. These specimens were dropped with an equal level of impact energy
as those in Chapter 2 to compare the patterns of ﬁ'acture. In Chapter 4, the data from
both impact scenarios was compared to assess the differences of fracture. A ﬁnite
element model was constructed to provide theoretical insight of the principal tensile
stress directions for better understanding of the fractures generated in each impact
scenario. The information presented in this thesis may be helpful diagnosing whether the

head was constrained in a forensic case.

ACKNOWLEDGMENTS

First and foremost, I would like to thank Dr. Roger Haut for his support and guidance in
my research and education over the past two years. Also, Dr. Todd Fenton for helping
me understand some basic concepts in forensic anthropology. I’d like to thank Cliff
Beckett for all of his technical assistance and Ed Reed and Star Lewis for supplying the
porcine specimens used for this research. I would like to thank Tim Baumer, M.S. and
Nicholas Passalacqua, M.S. for their dedication to the project as a whole. I would like to
thank my friends for all the laughs, support and good times through graduate school.
Finally, I’d like to thank my family for their unwavering support through the good times

and bad. I could not have done it without you!

iii

RESEARCH PUBLICATIONS BY THE AUTHOR

PEER-REVIEWED MANUSCRIPTS

Powell B, Passalacqua N, Baumer T, Fenton T, and Haut R. Fracture Patterns On the
Infant Porcine Skull Following Severe Blunt Impact. J. Forensic Sciences. (in
review)

Baumer T, Passalacqua N, Powell B, Newberry 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 TG, Powell BJ, Fenton TW, Haut RC, 2009, “Age Dependent Mechanical
Properties of the Infant Porcine Parietal Bone and a Correlation to the Human,”
Journal of Biomechanical Engineering, 131(11), pp. 111006-1-6.

PEER-REVIEWED ABSTRACTS

Powell BJ, Passalacqua NV, Baumer TG, Fenton TW, Haut RC. Fracture Patterns on the
Infant Porcine Skull Following Severe Blunt Impact. American Society of

Mechanical Engineers Summer Bioengineering Conference, Naples, Florida,
2010.

F enton TW, Passalacqua NV, Baumer TG, Powell BJ, Baumer TG, Newberry WN, Haut
RC. A Forensic Pathology Tool to Predict Pediatric Skull Fracture Patterns, Part
2: Fracture quantiﬁcation and further investigations on infant cranial bone
fracture properties. American Academy of Forensic Sciences (AAF S), Seattle,
Washington, 2010.

Van Wyhe RC, Powell BJ, Haut RC, Orth MW, Karcher DM. Reducing the growth rate
in turkeys improves femoral bone quality. International Poultry Expo, Atlanta,
GA, 2010.

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]

iv

TABLE OF CONTENTS

LIST OF TABLES .......................................................................................................... vi
LIST OF FIGURES ........................................................................................................ vii
CHAPTER 1: Developing the Porcine Model — An Introduction .................................... 1
CHAPTER 2: Fracture Patterns on the Infant Porcine Skull Following Severe
Blunt Impact ............................................................................................................. 10
Abstract .............................................................................................................. 10
Introduction ........................................................................................................ 12
Materials and Methods ....................................................................................... 13
Results ................................................................................................................ 17
Discussion .......................................................................................................... 23
References .......................................................................................................... 28

CHAPTER 3: Porcine Skull Fracture Length and Pattern Generated From a Free

Fall Impact ................................................................................................................ 31
Abstract .............................................................................................................. 3 1
Introduction ........................................................................................................ 33
Materials and Methods ....................................................................................... 34
Results ................................................................................................................ 38
Discussion .......................................................................................................... 43
References .......................................................................................................... 47

CHAPTER 4: Validation of Whole Head Drops in Free Fall Using a Porcine

Model ........................................................................................................................ 49
Abstract .............................................................................................................. 49
Introduction ........................................................................................................ 51
Materials and Methods ....................................................................................... 52
Results ................................................................................................................ 55
Discussion .......................................................................................................... 60
References .......................................................................................................... 64
CHAPTER 5: Conclusions and Recommendations for Future Work ............................ 66

References .......................................................................................................... 70
APPENDIX A: Raw Data ﬁom Chapter 2 ..................................................................... 71
APPENDD( B: Raw Data ﬁ'om Chapter 3 ..................................................................... 77

LIST OF TABLES

Table A. 1 . Raw data ﬁ'om high energy rigid interface impacts ..................................... 72
Table A2. Thickness, fracture length, and contact area measurements for high

energy rigid interface impacts ...................................................................... 74
Table A3. Raw data from high energy compliant interface impacts ............................. 75
Table A.4. Thickness, fracture length, and contact area measurements for high

energy compliant interface impacts .............................................................. 76
Table 3.]. Raw data collected ﬁ'om free fall rigid interface impacts ............................ 78

Table 3.2. Fracture length and contact area measurements for free fall high
energy rigid interface impacts ...................................................................... 8O

LIST OF FIGURES

Figure 1.1. Anatomy of the infant (top) and adult (bottom) human skull ........................ 2
Figure 2.1 . Orientation of the right parietal bone with rigid impact interface ............... 14
Figure 2.2. The drop tower schematic with GAM shown .............................................. 15

Figure 2.3. Peak impact force versus age for both the rigid and compliant
interfaces ........................................................... ‘ ........................................... 1 8

Figure 2.4. Contact area as a function of age for both rigid and compliant

interfaces ...................................................................................................... 18
Figure 2.5. The average length of skull fractures as a function of age .......................... 19
Figure 2.6. GIS map of 2-9 day old rigid (a) and compliant (b) impacts at high

energy ........................................................................................................... 20
Figure 2.7. GIS map of the 19-28 day old rigid (a) and compliant (b) impacts at

high energy ................................................................................................... 21
Figure 2.8. GIS map of the 2-9 day old rigid (a) and compliant (b) impacts at low

energy ........................................................................................................... 22
Figure 2.9. GIS map of the 19-28 day old rigid (a) and compliant (b) impacts at

low energy .................................................................................................... 23
Figure 3.1. Schematic of the ﬁ'ee fall drop tower ........................................................... 35

Figure 3.2. The drop trolley was raised to the necessary drop height and held with
the electromagnetic solenoid clamp (a). The free fall impact was
produced by disengaging the solenoid clamps (b). The mounting rod
was free to move vertically until the impact force returned to zero. ............ 36

Figure 3.3. An overlay of force-time plots showing the characteristic rapid drop in

force associated with skull fracture .............................................................. 39
Figure 3.4. Peak impact force with respect to age .......................................................... 39
Figure 3.5. Total fracture length with respect to age for free fall and entrapped

impacts .......................................................................................................... 40
Figure 3.6. The average total fracture length versus age ................................................ 4O

vii

Figure 3.7. GIS map of the 2-9 day old age group for the ﬂee fall (a) and
entrapped (b) impacts ................................................................................... 41

Figure 3.8. GIS map of the 10-17 day old age group for the ﬂee fall (a) and
entrapped (b) impacts ................................................................................... 42

Figure 3.9. Fracture initiation sites located remote of the point of impact
(represented by the bull’s—eye). Shaded areas represent ﬂactures and
each mark represents 5 mm of ﬂacture length ............................................. 43

Figure 4.1. Simpliﬁed skull geometry used for ﬁnite element simulations. Rigid
impactor positioned above the skull ............................................................. 53

Figure 4.2. Total ﬂacture (diastatic and bone) length with respect to specimen age
for both impact scenarios ............................................................................. 55

Figure 4.3. Peak impact force with respect to age for both ﬂee fall and entrapped
impacts .......................................................................................................... 56

Figure 4.4. Force versus time plots used to determine the duration of impact for
both entrapped and ﬂee fall impact scenarios .............................................. 56

Figure 4.5. Impact duration with respect to age for the ﬂee fall and entrapped
impacts .......................................................................................................... 57

Figure 4.6. Entrapped simulation showing impact site (bull’s-eye) and
surrounding principal tensile stress directions. Four primary areas of
maximum principal tensile stress were documented. Darker arrows
represent higher magnitudes of tensile stress. .............................................. 58

Figure 4.7. Overlaid maximum tensile stress magnitudes and directions on a
typical pattern of ﬂactures ﬂom an experimental entrapped porcine
specimen. Shaded areas indicated documented ﬂacture with each line
representing 5 millimeters ............................................................................ 58

Figure 4.8. Free fall simulation showing impact site (bull’s-eye) and surrounding
principal tensile stress directions. The largest magnitude of tensile
stress was located near the point of impact .................................................. 59

Figure 4.9. Overlaid maximum principal tensile stresses on an experimental ﬂee
fall porcine skull ........................................................................................... 59

Figure 4.10. Gurdjian’s (1975) comparison of the degree of head trauma resulting

ﬂom a blunt impact to the head. The study suggest that longer duration
impacts increase susceptibility of the skull to ﬂacture ................................. 61

viii

CHAPTER ONE

DEVELOPING THE PORCINE MODEL — AN INTRODUCTION

Head injury is the leading cause of death and permanent disability in children
(Tabatabaei and Sedighi, 2008). It is difﬁcult, however, to distinguish inﬂicted physical
abuse ﬂom accidental falls due to both scenarios producing similar types of head injuries
(Billmire and Myers, 1985). Linear, complex, and depressed skull ﬂactures have been
documented in both cases (Reece and Sege, 2000; Wheeler and Shope, 1997). Indeed,
skull ﬂacture is diagnosed in 1 of 3 children who are investigated for physical abuse
(Belfer et al., 2001). However, non-fatal injury resulting in skull ﬂacture can occur in
children less than 12 months of age ﬂom a fall ﬂom less than 3 feet (Gruskin and
Schutzman, 1999). In a study by Belechri et al. (2002), 1,881 cases of falls ﬂom bed
height reported no fatalities. Attempts have been made in developing a model for
pediatric trauma. One such model that has been met with some success scales adult head
impact data to predict the impact response of the pediatric head (Prange et al., 2004).
However, this model may be unreliable due to the head of a child being much smaller
and geometrically different than an adult (Schneider et al., 1986). Also, little information
is available on the similarity of skull fracture patterns between adults and children.

Predicting trauma in infants using adult data is also problematic due to the
signiﬁcantly different mechanical properties and structural characteristics of the
pediatric skull (Coats and Margulies, 2006). Fetal cranial bone is a very thin, non-
homogeneous and highly curved material (McPherson and Kriewall, 1980) while adult
bone is a three layered sandwich structure; two cortical bone plates surrounding a porous

diploé center (Motherway et al., 2009). In addition, several soft tissue junctions (sutures)

connect skull bones and allow growth of the brain during development. These sutures
begin to ossify and interdigitate into adulthood and eventually fuse together, making the
skull a single solid structure (Figure 1.1). The posterior and anterior fontanel are also
present in the infant skull. These membranous tissues also allow expansion of the brain
during development but eventually disappear after the skull bones fuse together. The
posterior fontanel is typically closed between birth and eight weeks of age, while the

anterior fontanel closes between nine and 26 weeks after birth (Knight, 1991).

 

Anterolateral .
fontanel Anterior
fontanel
Posterolateral ‘
fontanel
Posterior
Occipital bone Temporal fontanel
Coronal

 
  

SUIUIC

  
 
  

Frontal .

  

" Parietal

   

Squamous
suture ' ,3: {v . , . Coronal
‘_ Temporalh l .- a suture
Lambdoid .. " ‘ ‘
suture ’
, , Sagittal
Occrpital bone suture

 

Figure 1.1. Anatomy of the infant (top) and adult (bottom) human skull.
Because of ethical considerations, testing with infant humans is difﬁcult. There is
a current lack of studies that utilize infant cadaver specimens to produce data on the
impact response of the pediatric head. Infant anthropomorphic surrogates have been used
in a recent study (Prange et al., 2003) to document the rotational loading and

deceleration conditions of pediatric head impacts. While these studies produce valuable

quantitative data, there are no parts of these surrogates that fail at a given load to
simulate bone ﬂacture and the patterns of ﬂacture that are generated during impact
(Ewing et al., 1983). Animal models are commonly used for biomechanics studies to
generate correlative data to humans. The porcine animal model, in particular, has been
used ﬂequently in recent studies. Margulies and Thibault (2000) used porcine skull
samples to document age related changes in the mechanical properties of skull bone and
suture in three-point bending. Baumer et al. (2009) performed similar studies using four-
point bending of porcine skull beam specimens, documenting a skull development
relationship of days in pigs to months in infant humans.

In another study, Baumer et al. (In press) impacted the parietal bone on entrapped
infant porcine heads using a gravity accelerated mass with rigid and compliant
interfaces. The authors document ﬂaeture initiation occurring at the bone-suture
boundary, remote ﬂom the location of impact. Another key ﬁnding was that a compliant
interface generated more ﬂacture than a rigid interface for a given impact energy. The
study, however, was limited to only one, low level of impact energy. It has been
thoroughly shown in the literature that the height of a fall (impact energy) has an effect
on the degree of injury (Bertocci and Pierce, 2006; Barlow et al., 1983; Wilkins, 1997).
A fall ﬂom above 1.5 meters increases the severity of injury two fold, and a fall ﬂom
above 2.5 meters increases the severity of injury three fold (Macarthur et al., 2000;
Chalmers et al., 1996). In the process of developing a more robust pediatric ﬂacture
model, the effect of impact energy on the length of skull ﬂactures and the pattern of

ﬂacture must be examined.

Additionally, the impact scenario must be investigated. Often children sustain
head trauma ﬂom accidental falls during the development of walking motor skills
(Zimmerman and Bilaniuk, 1994). Death ﬂom falls is the third leading cause of death in
children aged 1-4 years (Hall et al., 1989). In the Baumer et al. (In press) study, the
porcine head model was entrapped in a bed of air-hardened epoxy, preventing translation
of the head after being impacted with a mass. In a study by Chason et al. (1966), a
canine head model that was both ﬁxed and ﬂee to move during impact was struck with a
rotary mass to assess the effect of head constraint on the degree of concussion. The
authors document no skull ﬂacture in the canine specimens, however, they did indicate
that there was a higher susceptibility to concussive and neurological problems when the
head was ﬁxed during impact. It has been shown in the literature, however, that less
force is typically required to produce a concussion than a skull ﬂacture (Rosman, 2004),
suggesting that the impact energy levels used by Chason et al. (1966) were not high
enough to produce skull ﬂacture. It is therefore unclear what effect constraint of the head
has on the degree of skull ﬂacture produced by a blunt impact. Investigation of this issue
is also necessary to generate a more robust model of pediatric skull ﬂacture.

In addition to animal models, Finite Element Analysis (FEA) is a powerful and
commonly used tool in biomechanics studies. While ﬂacture simulations have extremely
high computational costs, it has been noted in the literature that principal stress or strain
directions found in ﬁnite element simulations correlate well with experimentally and
clinically observed ﬂactures (Bozic et al., 1994; Silva et al., 1998). Frank and Lawn
(1967) propose that ﬂacture propagates perpendicular to the direction of the greatest

principal tensile stress as a method of dissipating the maximum amount of energy in a

system. Thus, it is assumed that the maximum principal stress directions may be used as
a predictor of ﬂacture in a ﬁnite element simulation.

This thesis presents research conducted on an in situ porcine head model, as well
as several ﬁnite element simulations to assess the effect of impact energy and impact
scenario on the degree and pattern of skull ﬂacture. In Chapter 2, ﬂacture length due to a
high level of blunt impact energy is documented and compared to previous impacts by
Baumer et al. (In press) at lower impact energy levels. Also, the patterns of ﬂacture for
the two studies are directly compared in the current study. This study hypothesized that
ﬂacture length would be a function of impact energy, specimen age, and impact
interface, and that the pattern of ﬂacture would change with a higher level of impact
energy. In Chapter 3, a ﬂee fall drop tower was built and used to assess the degree of
skull ﬂacture and the ﬂacture pattern for a ﬂee falling head onto a rigid interface. The
ﬂee fall ﬂacture patterns were then compared to those in Chapter 2. It was hypothesized
that the ﬂacture patterns for these two impact scenarios would be similar. In Chapter 4,
the total ﬂacture length, peak impact force, and impact duration data ﬂom Chapters 2
and 3 were compared and used in conjuncture with a ﬁnite element model to investigate
the impact responses of a rigid mass striking a constrained skull and a ﬂee falling head
onto a rigid surface. The maximum principal stress magnitudes and directions were
compared and used to discuss the experimental ﬂacture patterns.

The research conducted for this thesis provides insight into the ﬂacture patterns
of the developing porcine skull ﬂom two different impact variables: impact energy and
impact constraint. A ﬁnite element model was also used to validate the experimentally

observed differences in the degree of ﬂacturing. These data may ultimately aid in being

able to determine whether the head in a forensic case was constrained or ﬂee to move
during impact. This information may ultimately prove to be extremely beneﬁcial in
helping forensic pathologists and medical examiners to determine whether a trauma was

due to accident or abuse.

REFERENCES

Barlow B, Niemirska M, Gandhi RP, Leblanc W, 1983, “Ten years of experience with
falls ﬂom a height in children,” Journal of Pediatric Surgery, 18(4), pp. 509-511.

Baumer TG, Passalacqua NV, Powell BJ, Newberry WN, Fenton TW, Haut RC, In
Press, “Age-Dependent Fracture Characteristics of Rigid and Compliant Surface
Impacts on the Infant Skull — A Porcine Model,” Journal of Forensic Science.

Baumer TG, Powell BJ, Fenton TW, Haut RC, 2009, “Age Dependent Mechanical
Properties of the Infant Porcine Parietal Bone and a Correlation to the Human,”
Journal of Biomechanical Engineering, 131(11), pp. 111006-1-6.

Belechri M, Petridou E, Trichopoulos D, 2002, “Bunk versus conventional beds: a
comparative assessment of fall injury risk,” Journal of Epidemiology and
Community Health, 56(6), pp. 413-417.

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(2), pp. 122-4.

Bertocci G and Pierce MC, 2006, “Applications of Biomechanics Aiding in the
Diagnosis of Child Abuse,” Clinical Pediatric Emergency Medicine, 7(3), pp. 194-
199.

Billmire ME and Myers PA, 1985, “Serious head injury in infants: accident or abuse?”
Pediatrics, 75, pp. 340-342.

Bozic KJ, Keyak JH, Skinner HB, Bueff UH, Bradford DS, 1994, “Three-Dirnensional
Finite Element Modeling of a Cervical Vertebra: An Investigation of Burst Fracture
Mechanism,” Journal of Spinal Disorders & Techniques, 7(2), pp. 102-110.

Chalmers DJ, Marshall SW, Langley JD, Evans MJ, Brunton CR, Kelly A, Pickering
AF, 1996, “Height and surfacing as risk factors for injury in falls ﬂom playground
equipment: a case-control study,” Injury Prevention, 2(2), pp. 98-104.

Chason JL, Fernando 0U, Hodgson VR, Thomas LM, Gurdjian ES, 1966,
“Experimental Brain Concussion: Morphologic Findings and a New Cytologic
Hypothesis,” The Journal of Trauma, 6(6), pp. 767-779.

Coats B and Margulies SS, 2006, “Material Properties of Human Infant Skull and Suture
at High Rates,” Journal of Neurotrauma, 23(8), pp. 1222-1232.

Ewing CL, Thomas DJ, Sances A, Larson SJ, eds., 1983, “Impact Injury of the Head and
Spine,” lst Ed., Springﬁeld, IL: Charles C Thomas.

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

Gruskin KD and Schutzman SA, 1999, “Head Trauma in Children Younger Than 2
Years,” Archives of Pediatrics and Adolescent Medicine, 153, pp. 15-20.

Hall JR, Reyes HM, Horvat M, Meller J L, Stein R, 1989, “The Mortality of Childhood
Falls,” The Journal of Trauma, 29(9), pp. 1273-1275.

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

Macarthur C, Hu X, Wesson DE, Parkin PC, 2000, “Risk factors for severe injuries
associated with falls ﬂom playground equipment,” Accident Analysis and
Prevention, 32(3), pp. 377-382.

Margulies SS and Thibault KL, 2000, “Infant Skull and Suture Properties:
Measurements and Implications for Mechanisms of Pediatric Brain Injury,” Journal
of Biomechanical Engineering, 122(4), pp. 364-371.

McPherson GK and Kriewall TJ, 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.

Motherway JA, Verschueren P, Van der Perre G, Slotcn JV, Gilchrist MD, 2009, “The
mechanical properties of cranial bone: The effect of loading rate and cranial
sampling position,” Journal of Biomechanics, 42, pp. 2129-2135.

Prange MT, Coats B, Duhaime A, Margulies SS, 2003, “Anthropomorphic simulations
of falls, shakes, and inﬂicted impacts in infants,” Journal of Neurosurgery, 99(1), pp.
143-150.

Prange MT, Luck JF, Dibb A, Van Ee CA, Nightingale RW, Myers BS, 2004,
“Mechanical Properties and Anthropometry of the Human Infant Head,” Stapp Car
Crash Journal, 48, pp. 279-99.

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

Rosman NP, 2004, “Oski’s Essential Pediatrics,” In: M Crocetti and MA Barone,
editors. Oski’s Essential Pediatrics. Philadelphia: Lippincott, Williams, and Wilkins.

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 MJ, Keaveny TM, Hayes WC, 1998, “Computed Tomography-Based Finite
Element Analysis Predicts Failure Loads and Fracture Patterns for Vertebral
Sections,” Journal of Orthopaedic Research, 16(3), pp. 300-8.

Tabatabaei SM and Sedighi A, 2008, “Pediatric Head Injury,” Iranian Journal of Child
Neurology, 2(2), pp. 7-13.

Wheeler DS and Shope TR, 1997, “Depressed skull ﬂacture in a 7-month old who fell
ﬂom bed,” Pediatrics, 100, pp. 1033-1034.

Wilkins, B, 1997, “Head injury — abuse or accident?” Archives of Disease in Childhood,
76(5), pp. 393-397.

Zimmerman RA and Bilaniuk LT, 1994, “Pediatric head trauma,” Neuroirnaging Clinics
of North America, 4(2), pp. 349-366.

CHAPTER TWO

FRACTURE PATTERNS ON THE INFANT PORCINE SKULL
FOLLOWING SEVERE BLUNT IMPACT

ABSTRACT

Traumatic injury to the head accounts for 80% of fatal child abuse cases.
However, it is difﬁcult to distinguish abuse ﬂom accidental injury is difﬁcult due to
similarities in the type of injury. In this chapter, porcine specimens were impacted with a
rigid mass at a high level of energy to assess characteristic ﬂacture patterns related to
impact interface, specimen age, and level of impact energy. A single impact was
delivered to the right parietal bone of 57 specimens aged 2 to 28 days. The impact was
generated by releasing a gravity accelerated mass ﬂom a controlled height. Paired rigid
and compliant impacts of equal energy were conducted for each specimen age. Impact
force and contact area were recorded for each impact. Also, skull ﬂacture length was
measured to the nearest millimeter. Geographic Information Systems software was used
to monitor the ﬂequency of ﬂacture initiation and propagation. Individual maps were
generated based on impact interface, specimen age, and impact interface. It was found
that impact force increased with age, regardless of interface. Contact area increased with
age but was signiﬁcantly higher in the compliant interface impacts than the rigid. Skull
ﬂacture length was greater for the rigid interface than the compliant at all ages except for
2 days. The degree of skull ﬂacture also increased with an increase in impact energy.
GIS maps documented additional sites of ﬂacture initiation at a high level of impact
energy dependant upon impact interface and level of impact energy. Several unique

characteristic ﬂacture patterns were noted as a function of specimen age, impact

10

interface, and impact energy. These characteristics may prove to be extremely beneﬁcial
in establishing the causation of trauma in child abuse cases where the validity of the

testimony is questionable or unclear.

ll

INTRODUCTION

Head injuries account for 80% of fatal child abuse in young children (Case et al.,
2001). In a study of 89 children under the age of 2 years, 19 of the 20 fatalities were due
to abuse (Hobbs, 1984). In contrast, short falls rarely cause serious injury or death in
young children (Reiber, 1993). A study of 1,881 falls ﬂom bed height reported no deaths
(Belechri et al., 2002). Short falls (less than 3 feet), however, can still produce skull
ﬂacture in infants less than 12 months of age (Gruskin and Schutzrnan, 1999). There is
currently a lack of data regarding pediatric skull ﬂacture tolerance data and, thus,
pediatric trauma with related cranial ﬂacture due to a single-event represents one of the
greatest challenges to forensic pathologists and anthropologists.

Distinguishing between accidental and abusive trauma can be difﬁcult, as both
may produce similar types of injuries (Billmire and Myers, 1985). Speciﬁcally, linear,
complex, and depressed skull ﬂactures have been seen in both cases (Reece and Sege,
2000; Wheeler and Shope, 1997). The most commonly ﬂactured cranial bone in
accidental and abuse cases is the parietal (Hobbs, 1984; Meservy et al., 1987; Leventhal
et al., 1993). The risk of injury is also dependent on the contacting surface (Bertocci et
al., 2003). Other variables, such as the area struck, thickness of the skull, thickness of the
scalp and hair, and impact direction can also affect the pattern of skull ﬂacture (Knight,
1991; Cooperrnan and Merten, 2001). To better distinguish pediatric abuse ﬂom
accident, the aforementioned variables and the magnitude in which they affect skull
ﬂacture must be better understood.

A recent study by Baumer et al. (In press) assessed the effects of interface and

age using an infant porcine skull impact model, looking speciﬁcally at the location of

12

ﬂacture initiation on the parietal bone. The study shows that impacts ﬂom Iow heights
(low energy) typically initiate ﬂactures at a bone-suture boundary. However, in many
pediatric death cases there are multiple skull ﬂactures which sometimes extend across
suture boundaries (Meservy et a1, 1987; Stewart et al., 1993). Multiple, wide or cross-
suture ﬂactures are indicative of high energy trauma (Hobbs, 1984). It is therefore
necessary to study the effect of high energy impacts as a function of age and interface on
the ﬂacture patterns for the pediatric skull.

There were two hypotheses in the current study. First, that the locations of
ﬂacture initiation would not depend on the level of impact energy. Rather, it would only
depend on age and interface. Secondly, that an increase in impact energy would increase
the amount of ﬂacture via propagation for all ages and for both rigid and compliant
interfaces. These data will ultimately provide insight into the effect of impact energy on
skull ﬂacture patterns due to a single, blunt impact and may provide valuable
information that may help distinguish pediatric abuse ﬂom accident.

MATERIALS AND METHODS

Porcine specimens were received ﬂom a local supplier and stored at -20°C. A
total of 57 specimens (aged 2 to 28 days) were used for this study. The animals died of
natural causes and were ﬂozen within 12 hours of death. All specimens were ﬂee of
head injury, which was conﬁrmed during preparation.

The test procedure was described in a previous study (Baumer et al., in press).
Brieﬂy, the head was allowed to thaw at room temperature for 24 hours before the scalp
and facial tissues on the left side were removed. The specimens were transversely and

rotationally constrained in a bed of air-hardened epoxy (Fibre Strand, Martin Senour

13

Corp., Cleveland, OH). Phosphate-buffered saline (PBS) solution was applied regularly
during the preparation. The specimen was placed in a four degree of ﬂeedom ﬁxture
that allowed adjustments of the impact site (Figure 2.1). The skull was oriented such

that the center of the right parietal bone was normal to the impact interface.

Impact

Pressure Head

Film

Epoxy

Bed

Positioning
Fixture

 

Figure 2.1. Orientation of the right parietal bone with rigid impact interface.

The specimens were impacted using a gravity accelerated mass (GAM) (Figure
2.2). A single impact was delivered using an operational ampliﬁer comparator circuit to
monitor the impact force and energize an electromagnetic solenoid to catch the GAM
immediately after the impact force returned to zero. The forces were recorded using a
load transducer (4.45 kN capacity, model AL311CV, Sensotec, Columbus, OH) mounted
immediately behind the impact interface. Two interfaces were used in this study: rigid
and compliant. The rigid interface was a solid aluminum cylinder with approximately
16 cm2 of surface area (Figure 2.1). The compliant interface was a deformable aluminum
block (1.10 MPa crush strength Hexcel, Hexcel Corp, Stamford, CT), approximately 3

cm thick with a 16 cm2 surface area, attached to the rigid interface.

14

Stainless Steel Shaft

   
 
 
 
 
 
 
 
 
 

Vertically mounted
trolley with adjustable
mass (GAM)

Impact Head

Skull with parietal bone
exposed to apparatus

 

Positioning ﬁxture

Figure 2.2. The drop tower schematic with GAM shown.

Impact energy was controlled by varying the drop height of a 1.67 kg GAM. A
slightly larger, 1.92 kg mass, was used to generate ﬂacture in specimens aged twenty-
one days and older. The mass of the impact interface was included in the GAM. Energy
levels for each age were double those of a previous study (Baumer et al., in press). The
impact energy was doubled by raising the height of . the GAM to twice the height of the
previous study. The input energy for the compliant and rigid interfaces was equal at each
age, however, the impact energy was increased with specimen age. For cases in which
the ﬁrst impact did not cause ﬂacture of the skull (n=5 in the current study), the skull
was impacted a second time at a slightly higher energy. The force data was sampled at
10,000 Hz.

Pressure sensitive ﬁlm packets (Prescale, Fuji Film Ltd., Tokyo, Japan) were
attached to the impact site of each specimen to capture contact area. Two sheets of
polyethylene were used to protect low (0-10 MPa) and medium (10-50 MPa) range

pressure ﬁlms stacked on top of one other ﬂom ﬂuids (Atkinson et al., 1998). The

15

medium pressure ﬁlm was not used in the current study as the impact pressures were too
low to record on the ﬁlm.

Aﬂer impact, the remaining periosteum and soﬂ tissues were removed ﬂom the
skull and it was visually inspected for bone ﬂacture and suture damage. The remaining
soft tissue on the skull was then removed using standard anthropological procedures.
The length of skull ﬂactures was measured to the nearest millimeter using a soft, ﬂexible
measuring tape, which contoured to the curvature of the skull. Complete ﬂacture
diagrams were constructed manually for each specimen.

In order to compare the patterns of ﬂacture between specimens and interfaces, a
Geographic Information System (GIS) method was utilized in the study. The pattern of
ﬂacture ﬂom each skull was constructed using a projected view of the porcine cranium
which best highlighted the right side of the skull with ﬂacture conﬁgurations
superimposed on it for each specimen. A second view of the posterior aspect of the
cranium was also included as many high energy ﬂactures involved the occipital bone.
Fracture data ﬂom Baumer et. al (In press) was also revisited to compare low energy
rigid and compliant ﬂacture conﬁgurations to the current study. Porcine specimens were
separated into two different age groups (2-9 and 19-28 days) for both the rigid and
compliant impact interfaces and at low and high energy levels to better demonstrate
ﬂacture pattern changes in relation to porcine growth and development, impact interface,
and input energy. These age groups were chosen based on general observations of gross
ﬂacture and material property changes for the skull and suture tissues documented in the
literature (Baumer et al., in press; Baumer et al., 2009). The ﬂacture pattern for each

porcine cranium was traced into individual shape ﬁles (Marean et al., 2001). The GIS

16

model then counted overlaid ﬂacture patterns on each cranium, generating a map of
where ﬂactures appeared most ﬂequently. After each map was constructed, the GIS
model was used to discuss the differences in ﬂacture patterns between specimens of
different age, impact energy, and interface.

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

RESULTS

Impact energies were doubled ﬂom those used by Baumer et al. (In press) by
doubling the drop height at each age. The drop heights ranged ﬂom 0.2 m for a 2 day old
specimen to 1.2 m for a 28 day old specimen. The values of impact energy ranged ﬂom
3.1 J to 22.6 J, respectively.

The impact force on the skulls increased with age for both interfaces, and there
was little difference in the peak impact force (within 100 N at a given age) between the
two interfaces (Figure 2.3). Linear regression analysis indicated a signiﬁcant increase in

impact force with age for the compliant (p<0.001) and the rigid interfaces (p=0.006).

l7

H Rigid E] -- £1 Compliant
O

2500 -

2000 -

1500 -

1000 ~

Impact Force (N)

 

500 ~

 

 

o T I I I I I
0 5 10 15 20 25 30

A99 (Days)
Figure 2.3. Peak impact force versus age for both the rigid and compliant interfaces.

The contact areas generated during impact were found to signiﬁcantly increase
(p=0.003) with age at a similar rate for both interfaces (Figure 2.4). A two-factor
AN OVA (age, interface) for the contact area showed a signiﬁcantly larger area of

contact generated with the compliant than rigid interface.

1600 ~ HRigid D—UCompliant

A1200 ~

m2

Contact Area (m

 

 

 

Age (Days)
Figure 2.4. Contact area as a function of age for both rigid and compliant interfaces.

18

The length of ﬂacturing (in bone and along sutures) versus age plot showed, on
average, a signiﬁcantly larger (p=0.034) amount of ﬂacturing for the rigid than the

compliant interfaces (Figure 2.5).

 

 

 

 

 

 

 

 

 

 

 

 

200 — IRigid DCompliant
A160 -
E
E 1
£120 -
a:
c:
o
.1
w .1
g 80
o
E '1

404

O 7 I r r r I I
2 3 4 5 6 7 8 910111213141516171819202428
Age (Days)

Figure 2.5. The average length of skull ﬂactures as a function of age.

The GIS ﬂacture maps conﬁrmed that the length of ﬂactures was greater for rigid
than compliant interface impacts for the younger age group in the current study (Figure
2.6a and 2.6b). For the compliant interface experiments the pattern maps showed
ﬂactures primarily appearing to initiate at 4 sites adjacent to sutures along the perimeter
of the parietal bone. However, for the rigid, there appeared to be numerous initiation
sites. There was also signiﬁcantly more diastatic ﬂacturing in the rigid than compliant

interface experiments, speciﬁcally along the coronal suture.

19

Frequency

 

Figure 2.6. GIS map of 2-9 day old rigid (a) and compliant (b) impacts at high energy.

In the older goup of specimens (19-28 days) more ﬂacturing was again
conﬁrmed for the rigid than compliant interface experiments (Figure 2.7a and 2.7b). Yet,
in these experiments, no diastatic ﬂactures were noted. Sites of ﬂacture initiation were
evident in the parietal bone along the coronal and lambdoid sutures for the compliant
interface experiments. Two of these sites were similar to those documented in the
younger age goup. These sites were also noted in the rigid interface impacts, however
there were more propagated ﬂactures with the rigid interface. Interestingly, in the
current study using high energy impacts to the parietal bone, signiﬁcant ﬂacturing was

also documented in the occipital bone for botln age goups and interfaces.

20

Frequency

 

Figure 2.7. GIS map of the 19-28 day old rigid (a) and compliant (b) impacts at high
energy.

The GIS maps of the revisited Baumer et al. (In press) data showed three primary
areas of ﬂacture irnitiation, regardless of interface. For the younger age goup (2-9 days
old), the compliant interface produced more ﬂacture of the skull than the rigid at the

same impact energy level (Figure 2.8a and 2.8b).

21

Frequency

 

Figure 2.8. GIS map of the 2-9 day old rigid (a) and compliant (b) impacts at low
energy.

For a low energy of impact there was little to no skull ﬂacture with the compliant
interface for the older age goup (Figure 2%). Two ﬂacture initiation sites were noted
along the coronal and lambdoid sutures. The rigid impacts produced more propagated
ﬂactures initiating at approximately the same locations as in the compliant interface

experiments (Figure 2.9a).

22

Frequency

 

Figure 2.9. GIS map of the 19-28 day old rigid (a) and compliant (b) impacts at low
energy.

There were many specimens in botln high energy age goups where ﬂactures
appeared in the occipital region. These ﬂactures were not present in the revisited
Baumer et al. (In press) data.

DISCUSSION

The current study focused on skull ﬂacture patterns under a high impact energy
as a function of both age and interface using a porcine model. It was hypothesized that a
high level of impact energy would not change the locations of ﬂacture irnitiation ﬂom
those documented previously in low energy experiments (Baumer et al., in press).
However, the locations of ﬂacture initiation were found to be a function of impact
energy. Baumer et al. (In press) document three primary sites of ﬂacture initiation for
botln interfaces at a low impact energy. While these sites were present in the current
study, new initiation sites emerged at the increased level of impact energy. It was also

hypothesized in the current study that there would be a geater amount of ﬂactuning via

23

propagation for these higher energy impacts. The amount of ﬂacture produced in the
current study was signiﬁcantly geater than that produced in the low energy experiments
conducted by Baumer et al. (In press).

An interesting ﬁnding in the Baumer et al. (In press) study was the equal amount
of ﬂacture produced by the rigid and compliant interfaces at approximately 18 days of
age for a given impact energy. Prior to 18 days, the compliant interface produced more
ﬂacture than the rigid interface, but thereafter there was less for the same impact energy.
Baumer et al. (In press) suggest that the compliant interface generated higher states of
stress near sutures. These stresses were high enough to produce diastatic ﬂactures and
therefore a larger amount of total ﬂacture with the compliant interface. In the current
study, however, the rigid interface produced more ﬂacture than the compliant interface
at all ages, except for the 2 day old specimens. This change may be attributable to
alterations in bone and suture sensitivities to rate of loading. It has been shown in the
literature that human bone and suture exhibit mechanical property sensitivities to loading
rate (Wood, 1971; Yoganandan et al., 1995; Motherway et al., 2009). Young soft tissues,
in particular, are more sensitive to changes in loading rate than older tissues (Haut,
1983). Margulies and Thibault (2000) also determined the material properties of young
porcine crarnial suture at two different rates of loading and document signiﬁcant
increases in rupture modulus, elastic modulus and rupture energy at the higher rate.
Therefore, at higher rates of loading, sutures may become more “brittle-like”, especially
in the younger aged specimens. The rigid interface may have produced more ﬂacture
due its smaller contact area, which produced higher impact stresses than the compliant

interface. Furthermore, with an increase in suture stiffness for these high rates of loading

24

in the current study higher stresses were likely transmitted across sutures to produce
ﬂacture in the occipital bone with botln interfaces. These results contrast with the lack of
occipital ﬂacture documented in the Baumer et al. (In press) study at low levels of
impact energy. These ﬁndings suggest that high levels of impact energy generate more
areas of ﬂacture remote to the impact site.

The geater degee of total skull ﬂacture in the current study was also due to new
sites of ﬂacture initiation with increased levels of impact energy. Baumer et al. (In press)
document three primary sites of ﬂacture initiation, regardless of interface (Figure 2.8a
and 2.8b) at low energy. These three sites were also documented in the current study,
however, there were one or more additional sites of ﬂacture irnitiation depending on
interface. One possible explanation for these new sites is the need to dissipate a larger
amount of energy through ﬂacture of the bone. In the current study, the initiation sites
documented by Baumer et al. (In press) were fully propagated through the parietal bone
for the rigid interface. Fracture propagation appeared to be one method to dissipate
impact energy, however, additional ﬂactures sites were likely needed to allow the bone
to dissipate all of the impact energy generated with the higher drop heights in the current
study. These new sites were more ﬂequent with the rigid interface due to the impact
force being distributed over a smaller area, generating larger impact stresses in the skull.
The larger contact area produced with the compliant interface likely attenuated impact
stresses resulting in fewer ﬂacture initiation sites. This difference in ﬂacture initiation
sites suggests that the level of impact energy affected a characteristic feature of the

ﬂacture pattern for a given interface.

25

The commonality in ﬂacture patterns for the current and Baumer et al. (In press)
studies between specimens of the same age and impacted with the same interface was
documented using GIS software. This GIS image-analysis approach has been previously
used for both archaeological cut-mark distribution on fauna (Abe et al., 2002) and
carnivore modiﬁcation to fauna] remains (Hodgson et al., 2009), however, while
Darnann et al. (2009) have examined ﬂacture patterns ﬂom human aircraft crashes, this
is the ﬁrst forensic application of the Marean et al. (2001) GIS image-analysis technique
for bone ﬂacture pattern analysis. Furtlner, this project is unique as it is one of the ﬁrst
attempts to compile ﬂacture pattern data ﬂom a relatively large sample of documented
experimental impacts. Lee (1992) examined ﬂacture patterns of human laryngeal
structures after impacting tlnem with a drop-tower, but this analysis did not employ GIS,
instead using simple tracings of ﬂacture lines. The analyses in the current study provided
some insight into the discrimination of ﬂacture characteristics as a ﬁmction of specimen
age, impact interface and energy. These characteristics were described by assessing the
ﬂequency of ﬂacture on each GIS map for a given set of impact conditions. For
example, high energy impacts in the younger age goup (2-9 days) tended to produce
occipital ﬂacture for both interfaces. Again, occipital ﬂacture was not seen in the low
energy impacts. Additionally, each interface produced characteristic ﬂacture attributes.
The rigid interface generated much diastatic ﬂacturing at the higher impact energy,
whereas the compliant interface did not. These ﬁndings contrast with those noted by
Baumer et al. (In press) where the compliant interface produced more diastatic ﬂactures
than the rigid interface for the younger aged specimens. One could then say, if a given

ﬂacture pattern for a younger aged victim involves occipital and diastatic ﬂacture, the

26

causation of injury could have been due to a high energy, rigid impact. Future work
should focus on the study of uniqueness of the characteristic features of ﬂacture patterns
on the infant porcine skull as a function of impact interface, energy, and specimen age.
While there are certain age limitations of the porcine model (after approximately 24 days
the skull geometry begins to differ signiﬁcantly ﬂom humans), it could be used as an
experimental model to help develop a better understanding of these characteristics.

In cases of suspected child abuse, the medical examiner faces a difﬁcult task in
determining the causation of trauma. Age, interface and impact energy each appear to
affect the pattern and degee of skull ﬂacture. In the current study, rigid and compliant
interfaces were compared at a given specimen age under a high impact energy. The rigid
interface produced as much or more ﬂacture as the compliant at each age. This was in
contrast to previously reported data for low energy level impacts where a compliant
interface produced more ﬂacture of the infant porcine skull for specimens less than 18
days of age (Baumer et al., in press), and less for more aged specimens. The current
study also showed that a higher versus lower levels of impact energy in previous studies
altered the pattern of skull ﬂacture by generating new sites of ﬂacture irnitiation and
causing ﬂacture in adjacent bones of the skull. Some unique characteristics of these
ﬂacture patterns as a function of energy and interface were also assessed in the study
using GIS software. Development of these characteristics may prove to be extremely

beneﬁcial in the task of investigating cases of potential abuse to infants.

27

REFERENCES

Abe Y, Marean CW, Nilssen PJ, Assefa Z, Stone EC, 2002, “The analysis of cutrnarks
on archaeofauna: A review and critique of quantiﬁcation procedures, and a new
image-analysis GIS approach,” American Antiquity, 67(4), pp. 643-663.

Atkinson PJ, Newberry WN, Atkinson TS, Haut RC, 1998, “A method to increase the
sensitive range of pressure sensitive ﬁlm,” Journal of Biomechanics, 31(9), pp. 855-
859.

Baumer TG, Passalacqua NV, Powell BJ, Newberry WN, Fenton TW, Haut RC, in
press, “Age—Dependent Fracture Characteristics of Rigid and Compliant Surface
Impacts on the Infant Skull — A Porcine Model,” Journal of Forensic Science.

Baumer TG, Powell BJ, Fenton TW, Haut RC, 2009, “Age Dependent Mechanical
Properties of the Infant Porcine Parietal Bone and a Correlation to the Human,”
Journal of Biomechanical Engineering, 131(11), pp. 1-6.

Belechri M, Petridou E, Trichopoulos D, 2002, “Bunk versus conventional beds: a
comparative assessment of fall injury risk,” Journal of Epidemiology and
Community Health, 56, pp. 413-417.

Bertocci GE, Pierce MC, Deemer E, Aguel F, J anosky J E, Vogeley E, 2003, “Using Test
Dummy Experiments to Investigate Pediatric Injury Risk in Simulated Short-
Distance Falls,” Archives of Pediatrics and Adolescent Medicine, 157(5), pp. 480-
486.

Billrnire ME and Myers PA, 1985, “Serious head injury in infants: accident or abuse?”
Pediatrics, 75, pp. 340-342.

Case ME, Graham MA, Handy TC, Jentzen JM, Monteleone JA, 2001, “Position Paper
on Fatal Abusive Head Injuries in Infants and Young Children,” The American
Journal of Forensic Medicine and Pathology, 22, pp. 112-122.

Cooperman DR and Merten DF, 2001, “Skeletal Manifestation of Child Abuse,” In: RM
Reece & S Ludwig, editors. Child Abuse: Medical Diagnosis and Management.
Philadelphia: Lippincott, Williams, and Wilkins.

Darnann FE, Adler R, Benedix DC, Kontanis EJ, 2009, “Patterns of perimortern ﬂacture
ﬂom military aircraft crashes,” Proceedings of the American Academy of Forensic

Sciences; Washington DC.

Gruskin KD and Schutzrnan SA, 1999, “Head Trauma in Children Younger Than 2
Years,” Archives of Pediatrics and Adolescent Medicine, 153, pp. 15-20.

28

Haut RC, 1983, “Age-Dependent Inﬂuence of Strain Rate on the Tensile Failure of Rat-
Tail Tendon,” Journal of Biomecharnical Engineering, 105(3), pp. 296-299.

Hobbs CJ, 1984, “Skull Fracture and the Diagnosis of Abuse”, Archives of Disease in
Childhood, 59, pp. 246-252.

Hodgson JA, Plummer TW, Forrest F, Bose R, Oliver J S, 2009, “A GIS-based approach
to documenting large canid damage to bones,” Proceedings of the 78th annual
meeting of the American Association of Physical Antlnropologists; Chicago, IL.

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

Lee SY, 1992, “Experimental blunt injury to the larynx,” The Annals of Otology,
Rhinology, and Laryngology, 101(3), pp. 270-274.

Leventhal JM, Thomas SA, Rosenﬁeld NS, Markowitz RI, 1993, “Fractures in young
children: distinguishing child abuse ﬂom unintentional injuries,” American Journal
of Diseases of Children, 147, pp. 87-92.

Marean CW, Abe Y, Nilssen PJ, Stone EC, 2001, “Estimating the Minimum Number of
Skeletal Elements (MNE) in Zooarchaeology: A Review and a New Image-Analysis
GIS Approach,” American Antiquity, 66(2), pp. 333-348.

Margulies SS and Thibault KL, 2000, “Infant Skull and Suture Properties:
Measurements and Implications for Mechanisms of Pediatric Brain Injury,” Journal
of Biomecharnical Engineering, 122(4), pp. 364-371.

Meservy CJ, Towbin R, McLaurin RL, Myers PA, Ball W, 1987, “Radiogaphic
characteristics of skull ﬂactures resulting ﬂom child abuse,” American Journal of
Roentgenology, 149, pp. 173-175.

Motherway JA, Verschueren P, Van der Perre G, Sloten JV, Gilchrist MD, 2009, “The
mechanical properties of crarnial bone: The effect of loading rate and crarnial
sampling position,” Journal of Biomechanics, 42(13), pp. 2129-2135.

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

Reiber GD, 1993, “Fatal falls in childhood: How Far Much Children Fall to Sustain
Fatal Head Injury.” American Journal of Forensic Medicine and Pathology, 14, pp.
201-207.

Stewart G, Meert K, Rosenberg N, 1993, “Trauma in infants less than three months of
age,” Pediatric Emergency Care, 9(4), pp. 199-201.

29

Wheeler DS and Shope TR, 1997, “Depressed skull ﬂacture in a 7-month old who fell
ﬂom bed,” Pediatrics, 100, pp. 1033-1034.

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

Yoganandan N, Pintar FA, Sances A, Walsh PR, Ewing CL, Thomas DJ, et al., 1995,
“Biomecharnics of Skull Fracture,” Journal of Neurotrauma, 12(4), pp. 659-669.

30

CHAPTER THREE

FRACTURE LENGTH AND PATTERN OF THE PORCINE HEAD
MODEL IN FREE FALL

ABSTRACT

Head trauma is the leading cause of death in infants. Often, children sustain head
trauma during normal activities and play which are deemed accidental and innocent. In
the previous chapter, the degee and pattern of skull ﬂacture was investigated after a
porcine head model, entrapped in a bed of air-hardening epoxy, was struck with a blunt
mass. This scenario, however, does not reﬂect the common injury occurrence ﬂom
accidental falls. In the current study the porcine head model was dr0pped in ﬂee fall to
assess the pattern of ﬂacture generated ﬂom an unconstrained fall ﬂom a controlled
height. A mounting plate and clamp was used to orient the right parietal bone normal to
the rigid impact interface. A drop trolley translating along a vertical stairnless steel shaft
attached to a drop tower provided motion in only the vertical direction. Drop heights
were calculated ﬂom previously used impact energy values in Chapter 2 in order to
produce an equal amount of ﬂee fall impact energy to the skull. The Geogaphic
Information Systems method was again used to map the ﬂequency of fractures in the
parietal, ﬂontal and occipital bones. Total ﬂacture length steadily decreased with age in
the ﬂee fall head impacts. The results also showed that the pattern of ﬂacture differed
signiﬁcantly between the ﬂee fall impacts and entrapped impacts regardless of age. The
extensive levels of occipital bone ﬂacture documented in Chapter 2 was not seen in the
ﬂee fall impacts, even though the same impact energy was used. Additionally, there

were signiﬁcantly fewer sites of ﬂacture initiation in the ﬂee fall impacts than in the

31

entrapped impacts. These sites were age dependent with younger specimens having
ﬂacture initiation along the coronal suture and older specimens having ﬂacture initiation
sites on botln the posterior and anterior edges of the parietal bone along the coronal and
larndoid sutures. Diastatic ﬂacture was documented in both age goups but not in
specimens older than 14 days of age. The patterns of ﬂacture presented here are
signiﬁcantly different and are representative of two cases of equal high level of energy
head impacts. One case is a fall with a given energy onto a rigid surface and the second
case is a head that is on a rigid surface being struck with a blunt object. The research
presented in this chapter may be extremely beneﬁcial to medical examiners and forensic
pathologists in determining the cause of skull ﬂacture in cases of potential pediatric

abuse.

32

INTRODUCTION

Traumatic injury to the head is the leading cause of death in infant humans
(Tabatabaei and Sedighi, 2008). Often children sustain head trauma due to accidental
falls during the development of walking motor skills or ﬂom child support devices such
as highchairs (Zimmerman and Bilaniuk, 1994; Mayr et al., 1999). Pediatric falls
typically result in impacts to the head due to the increased weight ratio of the head to the
body as compared to adults (Y oganandan and Pintar, 2004; Snyder, 1977; Smith et al.,
1975; Cory et al., 2001). Death resulting ﬂom falls is the third leading cause of death in
infants aged 1-4 years of age (Hall et al., 1989). However, in a study of 89 children
under 2 years of age, 19 of the 20 fatalities were attributed to physical abuse (Hobbs,
1984). Linear, complex, and depressed ﬂactures have been documented in both abuse
and accidental cases (Reece and Sege, 2000; Wheeler and Shope, 1997). Thus,
distinguishing an accident ﬂom inﬂicted abuse is difﬁcult due to both scenarios
producing similar types of injury (Billmire and Myers, 1985). Due to a lack of pediatric
cranial trauma data, correctly diagnosing skull ﬂacture due to abuse or ﬂom an
accidental fall poses a signiﬁcant challenge to medical examiners and forensic
pathologists and anthropologists.

Injury biomechanics are used in case-based investigations of suspected child
abuse (Bertocci and Pierce, 2006). Animal models are often used in these cases to
correlate data to humans. A porcine head model has been recently used in this laboratory
to assess the effect of a blunt impact to the parietal bone of the skull. Baumer et al. (In
press) document ﬂacture initiation sites occurring at the bone-suture boundary, remote

of the point of impact. Similarly, in Chapter 2, skull ﬂacture patterns were assessed to

33

distinguish unique characteristics due to impact interface, age of specimen, and impact
energy. In both studies, the porCine head was entrapped in a bed of epoxy and impacted
with a rigid mass. Chason et al. (1966) impacted a carnine head model that was both
ﬁxed and ﬂee using a rotary striker. Interestingly, the authors documented that there was
a higher susceptibility to concussion when the head was ﬁxed rather than ﬂee to move.
The authors also documented a lack of skull ﬂacture in all of the specimens. It has been
shown, however, that there is typically less force required to produce a concussion than a
skull ﬂacture (Rosman, 2004) and thus the energy levels used in the Chason et al. (1966)
study may not have been high enough to produce skull ﬂacture. Thus, it is unclear what
effect constraint of the head during impact has on the degee and pattern of skull
ﬂacture. Therefore, in the current study, the porcine head model was dropped in ﬂee fall
to assess the degee of skull ﬂacture length as well as the ﬂacture pattern.

The hypothesis of this study was that the pattern of ﬂacture in the ﬂee fall impact
would be similar to those found in the entrapped impacts ﬂom Chapter 2 for an equal
impact energy. These data may ultimately provide utility for forensic pathologists and
anthropologists in comparing ﬂacture data ﬂom potential pediatric abuse cases to known
impact conditions using the porcine head model. If there are differences in fracture
pattern between a constrained head impact versus a ﬂee fall head drop, it may be
possible to determine if the victim’s head was constrained or ﬂee to move during the
impact.

MATERIALS AND METHODS
A total of 31 porcine specimens were received ﬂom a local supplier and stored at

a temperature of -20°C. All animals had died of natural causes and were ﬂozen within

34

12 hours of death. The specimens ranged ﬂom 2-17 days of age. Each animal was
inspected for initial head injury by palpating the parietal bone prior to experimentation.
Each specimen was fastened to a mounting plate using Velcro straps. A four
degee of ﬂwdom clamp directly attached to the mounting plate was used to orient the
parietal bone normal to the impact interface. The mounting plate was fastened to a
hollow aluminum rod which was supported by a gravity accelerated drop trolley. The
rod was clamped with an electromagnetic solenoid, acting as a catch and release
mechanism, attached to the drop trolley. The drop trolley was constricted to linear
translation in the vertical direction by a stairnless steel shaft attached to a drop tower
chassis (Figure 3.1). To produce the necessary impact energy, the drop trolley was raised
to the necessary drop height and held in place by a second electromagnetic solenoid. The
solenoid was fastened to a crossbar to adjust the drop height to a maximum of 9 feet.
During the experiment, both solenoids disengaged, allowing the drop trolley and

mounting rod to fall ﬂeely (Figure 3.2).

Stainless Steel Shaft

 
  
  

Vertically mounted
drop trolley

 

Mounting plate
with porcine head

Trolley Impact
Padding

 

Impact interface
and load cell

- ... _ ... . , ,.... _
V r I‘ ‘1 7.3.” v-7": r, ., .,, ,_
t if . .
J‘“n

~.— +1 «ate: ra'ricrjmx-i? .. ‘z I .~;-r-_ _ I r‘

Figure 3.1. Schematic of the ﬂee fall drop tower.

35

Upon impact, the trolley base struck a soft padded surface to dissipate the drop energy in
the trolley. The skull impacted a rigid, aluminum interface with an approximate surface
area of 324 cm2. A load cell (2.22 kN capacity, model IOIOAF-SOO, Interface,
Scottsdale, AZ) mounted immediately behind the impact interface recorded impact
force, duration, and energy. The skull was allowed to impact only once by using an
operational ampliﬁer comparator circuit to monitor the impact and reenergize the
electromagretic solenoid to catch the rod immediately after the impact force returned to
zero. The force data were sampled at 10,000 Hz.

Solenoid clamps Solenoid clamps
engaged disengaged

 

 

 

 

a b
Figure 3.2. The drop trolley was raised to the necessary drop height and held with the

electromagnetic solenoid clamp (a). The ﬂee fall impact was produced by disengaging
the solenoid clamps (b). The mounting rod was ﬂee to move vertically until the impact
force returned to zero.
In Chapter 2, entrapped porcine specimens were impacted using a high level of

impact energy. Energy values for each age of specimen were matched in this chapter to

those used in Chapter 2 by varying the drop height of each specimen. Due to the

36

differences in mass of the head with age, each specimen’s head mass was measured and
given a respective drop height to match the Chapter 2 impact energies at each age using

U = m * g * h (4.1)
where U was the potential impact energy, m was the mass of the head, g was the
gavitational acceleration, and h was the respective drop height.

Each skull was dropped only once regardless if skull ﬂacture was present. After
impact, the scalp and soft tissues of the skull were removed using standard
anthropological procedures. Each skull was visually inspected for ﬂactures and photo
documentation was taken. The periosteum and remaining soft tissues were then
removed. The length of skull ﬂacture was measured to the nearest millimeter using a
soft, ﬂexible measuring tape, which contoured to the curvature of the skull. Complete
ﬂacture diagarns were manually constructed for each specimen.

The Geo gaphic Information System (GIS) method was again used in this chapter
to map the ﬂequency of ﬂactures. The pattern of ﬂacture ﬂom each skull was
constructed using a projected view of the porcine cranium which best highlighted the
right side of the skull with superimposed ﬂacture conﬁgurations for each specimen. A
second view of the posterior aspect of the cranium was also included to display ﬂactures
of the occipital bone. The porcine specimens were again separated into two different age
groups (2-9 and 10-17 days) to demonstrate ﬂacture pattern changes in relation to
porcine gowth and development, as well as the impact scenario. Fracture pattern data
ﬂom Chapter 2 were revisited to compare the ﬂacture patterns of the 2-9 day old and 10-
17 day old age goups from high energy, rigid interface, entrapped impacts to the ﬂee

fall experiments. The age goups were chosen based on the available specimens, as well

37

as general observations of goss ﬂacture and material property changes for the skull and
suture tissues documented in the literature (Baumer et al., in press; Baumer et al., 2009).
The pattern of ﬂacture for each porcine cranium was traced into individual shape ﬁles
(Marean et al., 2001). The GIS software then counted overlaid ﬂacture patterns on each
crarnium, generating a map of where ﬂactures appeared most ﬂequently. After each map
was constructed, the GIS maps were used to compare the patterns of ﬂacture produced
ﬂom ﬂee falls and entrapped impacts using the same energy.

Impact force and total ﬂacture length were analyzed for age effects using linear
regession analyses in a statistics software (SignaStat 2.03, Aspire Software
International, Ashbum, VA). Statistically signiﬁcant effects were reported for a p-value
of less than 0.05.

RESULTS

The aim of the current study was to assess the impact characteristics of a porcine
head dropped in a ﬂee fall scenario. There was a characteristic rapid drop in force
documented in the force-tirne plots, which was associated with skull ﬂacture (Figure

3.3). A more pronounced peak was noted if no ﬂacture was present.

38

1400 -
1200 -

     

1000 i

 

Fracture
800 -

Force (N)

600 -
400 -
200 -

 

0 r . I

0 0.001 0.002 0.003
Impact Duration (seconds)

Figure 3.3. An overlay of force-time plots showing the characteristic rapid drop in force
associated with skull ﬂacture.

 

Peak impact force signiﬁcantly increased with age at a rate of 40.9 N/day of age

(p<0.001) (Figure 3.4). Also, impact energy signiﬁcantly increased with age at a rate of
0.19 J/day (p=0.009).

0 Impact Force - Impact Energy

 

 

 

 

1600 — —— 9
@1200 — 3
3 " 65
'5 ~ 2
I: 800 m
0 H
3 . . -- 3%
_§ 400 - . a I E

0 I . I I 0

o 5 1o 15 20

Age (Days)
Figure 3.4. Peak impact force with respect to age.

39

Total ﬂacture length (bone and diastatic) was found to signiﬁcantly decrease
with age at a rate of -l.83 mm/day (p<0.001) (Figure 3.5). In 5 of the 31 specimens, no

skull ﬂacture (bone or diastatic) was documented.

80~

O)
O
r
I

Total Fracture Length (mm)
B 8

 

 

 

O

 

10 15 20
Age (Days)

0
(1'1

Figure 3.5. Total ﬂacture length with respect to age for ﬂee fall and entrapped impacts.
Average diastatic and bone ﬂacture were compared at each age (Figure 3.6).
Diastatic ﬂacture was documented as early as 3 days of age. However, no diastatic

ﬂacture was noted in specimens older than 14 days of age.

0)
O
J

I Bone El Diastatic

.h
o
r

N
O
r

 

Average Fracture Length (mm)

 

2 3 4 5 6 7 8 91011121314151617
Age(Days)

Figure 3.6. The average total ﬂacture length versus age.

0
r

40

In the younger age goup (2-9 days old), extensive diastatic fracture was
documented along the coronal suture in the ﬂee fall impacts (Figure 3.73). Additionally,
several bilateral (occurring on both sides of the suture) bone ﬂactures extended across
the coronal suture into the ﬂontal and parietal bones. Bone ﬂacture tended to initiate at
the coronal suture. There were no documented initiation sites along the posterior or
superior edges of the parietal bone. Also, there was no documented occipital ﬂacture in

the ﬂee fall head impacts contrasting the extensive occipital ﬂacture noted in Chapter 2

(Figure 3.7b).

Frequency

CEO-1
-2-2
-3-4
-5-6

 

Figure 3.7. GIS map of the 2-9 day old age goup for the ﬂee fall (a) and entrapped (b)
impacts.

In the older age goup (10-17 days old), several ﬂacture initiations were
documented at the lamdoidal and squamosal suture intersection (Figure 3.83). However,
ﬂacture initiation was documented more ﬂequently along the anterior parietal bone as in
the younger age goup. In Chapter 2, the sites of ﬂacture initiation were ﬂequent and

numerous along the perimeter of the parietal bones (Figure 3.8b).

41

Frequency

 

Figure 3.8. GIS map of the 10-17 day old age goup for the ﬂee fall (a) and
entrapped (b) impacts.

As with the young age goup, no occipital ﬂacture was documented in the 10-17 day old
age goup when dropped in ﬂee fall. This again contrasted with the extensive occipital
ﬂacturing documented for the entrapped head impacts of Chapter 2. Diastatic ﬂacture in
the coronal suture was again recorded in the older age goup. However, the degee of
diastatic ﬂacture was signiﬁcantly less in the older age goup than in the younger.
Several of the older specimens had ﬂacture irnitiation on either side of the parietal
bone at locations remote ﬂom the point of impact (Figure 3.9). This was also
documented in Chapter 2 and in a previous study using entrapped porcine head rigid

impacts (Baumer et al., in press).

42

 

Figure 3.9. Fracture initiation sites located remote of the point of impact (represented by
the bull’s-eye). Shaded areas represent ﬂactures and each mark represents 5 mm of
ﬂacture length.

DISCUSSION

The GIS maps generated in the current chapter did not validate the hypothesis
that the pattern of skull ﬂacture was similar between free fall and entrapped head
impacts. In Chapter 2, signiﬁcant occipital ﬂacture was documented in both age groups
(2-9 days old and 10-17 days old) for entrapped head impacts at the high level of impact
energy. However, irn the current chapter the ﬂacturing generated ﬂom a ﬂee fall head
drop, with an equivalent level of impact energy, was confined to the parietal and frontal
bones. Thus, it might be hypothesized that the presence of occipital ﬂacture is
attributable to an increased state of overall skull stress when the skull is rigidly
constrained in the entrapped impacts. Further investigation into this claim will be
conducted in Chapter 4. Additionally, diastatic ﬂacture was localized to only the coronal
suture for all ages in the ﬂee fall impacts. In the entrapped impacts, diastatic was

documented extensively in all sutures surrounding the parietal bone for young ages. For

older aged specimens, diastatic ﬂacture was documented only in the coronal and sagittal

43

sutures. These differences further validate that the pattern of ﬂacture changed with
different impact scenarios.

In the 2-9 day old age goup, the ﬂacture patterns generated ﬂom a blunt impact
to the entrapped head produced ﬂequent and diffuse ﬂacture initiation sites along the
parietal bone boundaries. In the ﬂee falls, however, the ﬂacture initiation sites were
localized along the coronal suture. However, the two sets of ﬂacture patterns may be
used in tandem to diagnose if the head was constrained during the impact by comparing
the amount of fracture initiation sites of the victim to the data. The entrapped head
fracture patterns provide a better indicator of skull ﬂacture pattern generated ﬂom an
impact where the head may be resting against a rigid surface. An example of this may be
when a victim’s head is pinned against a concrete wall and is struck with a blunt weapon
such as a club. The ﬂacture patterns generated ﬂom the ﬂee fall head impacts would be
more representative of cases where the victim fell ﬂom a highchair or countertop or
ﬂom a trip and fall incident during the development of walking motor skills. If in a case
of potential child abuse, the victim has few initiation sites located along the coronal
suture, ﬂom the current study data, it can be reasoned that the trauma was due to some
sort of ﬂee falling head impact with a rigid surface. While abuse cannot be ruled out (the
child may have been pushed over), it does rule out a blow to the head while the child’s
head was constrained with an equal impact energy.

In the current study, the pig model was dropped onto the right parietal bone for
better comparison to the data in Chapter 2, which also involved impact trauma to the
right parietal bone. The two studies were conducted to record differences in ﬂacture

patterns between the two impact scenarios. The literature has shown that the most

44

commonly ﬂactured cranial bone in both accidental and abuse cases is the parietal
(Hobbs, 1984; Meservy et al., 1987; Leventlnal et al., 1993) and thus it was reasonable
that the targeted impact area of the skull in Chapter 2 and 3 should be the parietal.
However, it has been documented in the literature that accidental falls in children
typically result in vertex impacts to the head (Yoganandan and Pintar, 2004). This is
attributed to the larger head to body mass ratio of children than adults (Snyder, 1977). It
must be noted, however, that not all accidental head injuries occur in this fashion. In a
study of 18 children with cranial trauma (Plunkett, 2001), several children fell onto the
occipital region and also onto the parietal region in innocent playgound accidents. In
order for the medical examiner to validate the testimony of the caregiver, the pattern of
skull ﬂacture must be known for all impact conditions, including impact locations. In
this regard, the current study was limited to parietal bone impacts.

In the ﬂee fall impacts, ﬂacturing was typically linear and remote ﬂom the point
of impact. The ﬂactures in the young specimens were generally bilateral, spanning into
both the ﬂontal and parietal bones. However, in the older age goup, the ﬂactures were
unilateral and present in only the parietal bone. Hobbs (1984) documents that accidental
injury typically generated narrow, linear ﬂactures in the parietal bone. The data in this
chapter supports this claim but it is unclear why there are bilateral ﬂactures and
unilateral ﬂacture differences between the young aged and the older aged specimens. It
was suggested in Chapter 2 that higher rates of loading cause young suture tissue to
behave more “brittle-like”, allowing higher stress distributions into surrounding bones of
the skull. This same suggestion can be made for the young aged ﬂeely dropped

specimens in the current chapter. The impact stresses were distributed into the ﬂontal

45

bone causing ﬂacture because of the suture’s ability to transmit stress at higher rates of
loading. In the older aged specimens, the sutures have become more stiff and the skull
behaves as a single, complete structure rather than separate entities of bone. This allows
the impact stresses to be transmitted over the entire area of the skull reducing the
potential of ﬂacture.

The results of this chapter document that a porcine model dropped in ﬂee fall
produced a different pattern of ﬂacture, for a given energy, than an entrapped head
impacted with a blunt object. Interestingly, there was a lack of occipital bone ﬂacture in
the ﬂee fall cases, which was attributed in Chapter 2 to high levels of impact energy. It
was also found that the site of fracture initiation was age-dependent with older
specimens having ﬂacture initiation on opposite sides of the parietal bone. Accidental
falls are a common occurrence in child development and it is necessary to know the
pattern of ﬂacture generated ﬂom a ﬂee fall head impact. The information produced in
this research could prove extremely valuable to forensic pathologists and medical
examiners in distinguishing the cause of skull ﬂacture in a potential child abuse case

when the circumstances of injury are questionable or unknown.

46

REFERENCES

Baumer TG, Passalacqua NV, Powell BJ, Newberry WN, Fenton TW, Haut RC, in
Press, “Age-Dependent Fracture Characteristics of Rigid and Compliant Surface
Impacts on the Infant Skull — A Porcine Model,” Journal of Forensic Science.

Baumer TG, Powell BJ, F enton TW, Haut RC, 2009, “Age Dependent Mechanical
Properties of the Infant Porcine Parietal Bone and a Correlation to the Human,”
Journal of Biomechanical Engineering, 131(11), pp. 1-6.

Bertocci G and Pierce MC, 2006, “Applications of Biomechanics Aiding in the
Diagnosis of Child Abuse,” Clinical Pediatric Emergency Medicine, 7(3), pp. 194-
199.

Billmire ME and Myers PA, 1985, “Serious head injury in infants: accident or abuse?”
Pediatrics, 75, pp. 340-342.

Chason JL, Fernando OU, Hodgson VR, Thomas LM, Gurdjian ES, 1966,
“Experimental Brain Concussion: Morphologic Findings and a New Cytologic
Hypothesis,” The Journal of Trauma, 6(6), pp. 767-779.

Cory CZ, Jones MD, James DS, Leadbeatter S, Nokes LDM, 2001, “The potential and
limitations of utilizing head impact injury models to assess the likelihood of

signiﬁcant head injury in infants after a fall,” Forensic Science International, 123,
pp. 89-106.

Hall JR, Reyes HM, Horvat M, Meller JL, Stein R, 1989, “The Mortality of Childhood
Falls,” The Journal of Trauma, 29(9), pp. 1273-1275.

Hobbs CJ, 1984, “Skull Fracture and the Diagnosis of Abuse”, Archives of Disease in
Childhood, 59, pp. 246-252.

Leventhal JM, Thomas SA, Rosenﬁeld NS, Markowitz RI, 1993, “Fractures in young
children: distinguishing child abuse ﬂom unintentional injuries,” American Journal
of Diseases of Children, 147, pp. 87-92.

Marean CW, Abe Y, Nilssen PJ, Stone EC, 2001, “Estimating the Minimum Number of
Skeletal Elements (MNE) in Zooarchaeology: A Review and a New Image-Analysis
GIS Approach,” American Antiquity, 66(2), pp. 333-348.

Mayr JM, Seebacher U, Schimpl G, Fiala F, 1999, “Highchair accidents,” Acta
Paediatrica, 88(3), pp. 319-322.

Meservy CJ, Towbin R, McLaurin RL, Myers PA, Ball W, 1987, “Radiogaphic

characteristics of skull ﬂactures resulting ﬂom child abuse,” American J oumal of
Roentgenology, 149, pp. 173-175.

47

Plunkett J, 2001, “Fatal Pediatric Head Injuries Caused by Short-Distance Falls,” The
American Journal of Forensic Medicine and Pathology, 22(1), pp. 1-12.

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

Rosman NP, 2004, “Oski’s Essential Pediatrics,” In: M Crocetti and MA Barone,
editors. Oski’s Essential Pediatrics. Philadelphia: Lippincott, Williams, and Wilkins.

Smith MD, Burrington JD, Woolf AD, 1975, “Injuries in children in ﬂee falls: an
analysis of 66 cases,” Journal of Trauma, 15, pp. 987-991.

Snyder RG, F oust DR, Bowman BM, 1977, “Study of impact tolerance through ﬂee fall
investigations,” III-IS, Washington DC.

Tabatabaei SM and Sedighi A, 2008, “Pediatric Head Injury,” Iranian Journal of Child
Neurology, 2(2), pp. 7-13.

Wheeler DS and Shope TR, 1997, “Depressed skull ﬂacture in a 7-month old who fell
ﬂom bed,” Pediatrics, 100, pp. 1033-1034.

Yoganandan N and Pintar FA, 2004, “Biomechanics of temporo-parietal skull ﬂacture,”
Clinical Biomechanics, 19, pp. 225-239.

Zimmerman RA and Bilaniuk LT, 1994, “Pediatric head trauma,” Neuroirnaging Clinics
of North America, 4(2), pp. 349-366.

48

CHAPTER FOUR

COMPARISON OF FREE FALL AND ENTRAPPED IMPACTS WITH
FINITE ELEMENT ANALYSIS

ABSTRACT

Signiﬁcant differences exist in the pattern and degee of ﬂacture in head impacts
onto a rigid surface and impacts to a constrained head with a rigid interface. In this
chapter, ﬂacture length, peak impact force, and impact duration data ﬂom the previous
two chapters were compared to assess the differences between the two scenarios. Also,
ﬁnite element simulations were performed using a simpliﬁed skull geometry in order to
further investigate the differences ﬂom a theoretical viewpoint. Peak impact forces were
found to be similar regardless of the impact scenario. However, signiﬁcantly more
ﬂacture length was documented in the entrapped porcine head impacts than in the ﬂeely
dropped heads. Impact durations were signiﬁcantly longer in the entrapped impacts
leading to the assumption that a longer impact duration was the likely explanation for
more propagated ﬂactures. Several sources in the literature conﬁrm that a longer
duration impact increases the susceptibility of material ﬂacture. The maximum principal
tensile stresses were compared between the two impact scenarios using the ﬁnite
element model. The results showed that the stresses in the entrapped scenario were
distributed over a larger area of the skull and were geater in magnitude at locations
remote ﬂom the point of impact. The ﬂee fall simulation produced maximum principal
tensile stresses at or nearer to the point of impact. Overlays of the principal stresses onto
an experimentally impacted entrapped porcine head showed a strong correlation of

ﬂacture pattern between the theoretical ﬁnite element model and the experimental data.

49

However, the ﬂee fall principal stresses contrasted the experimental ﬂacture pattern in
which ﬂacture initiated at the bone-suture boundary and not at the point of impact which
the ﬁnite element results suggested. The model was limited, however, to a homogeneous
material and simple geometry. A more complex geometry with bone-suture interfaces
may provide a better insight into the ﬂacture patterns in the ﬂeely dropped porcine

heads.

50

INTRODUCTION

In the previous chapters, two different impact scenarios of the infant porcine
skull were tested. In Chapter 2, an entrapped porcine head was impacted with a blunt,
rigid mass. These impacts were conducted with a high level of impact energy and
generated extensive propagated ﬂactures in the parietal and adjacent bones of the skull.
The impact scenario was representative of a constrained head that was impacted with a
blunt object, most likely occurring in an inﬂicted situation. However, a majority of
pediatric head trauma is due to accidental falls (Kim et al., 2000; Hall et al., 1989), and
thus the porcine head model was tested in ﬂee fall in Chapter 3. Signiﬁcant differences
in the ﬂacture patterns were documented in the two studies. Since the level of impact
energy in the two studies was the same, the variation in ﬂacture patterns must have been
attributed to head constraint.

Often in biomechanics studies, Finite Element Analysis (F EA) is used to
determine material stresses and strains in complex geometries or dynarrnic situations.
Several studies have shown that ﬁnite element models can be used to correlate clinically
or experimentally observed ﬂacture patterns with the stress or strain distributions found
ﬂom ﬁnite element simulations (Silva et al., 1998; Doorly and Gilchrist, 2006; Baumer,
2009; Baumer et al., in press). Modeling ﬂacture of a material typically requires a large
computing power, but Frank and Lawn (1967) proposed a theory that principal stress
directions could be used to predict the path of ﬂacture propagation. They propose that
ﬂacture, as a means of dissipating the maximum amount of energy in a system, occurs

perpendicular to the maximum principal stress direction. Therefore, it may be possible to

51

predict areas of a model that are more susceptible to ﬂacture by investigating the
principal stress directions.

In this chapter, a comparison was made between the entrapped head and ﬂee fall
impact data ﬂom the previous two chapters. Fracture length, impact duration, and peak
impact force data were revisited ﬂom Chapters 2 and 3. Also, ﬁnite element analysis
was used to simulate the two impact scenarios using a simpliﬁed skull geometry. It is
thought that by assessing the principal stress magnitude and directions found in the ﬁnite
element analysis, the change in ﬂacture pattern between entrapped and ﬂeely dropped
heads could be explained.

MATERIALS AND METHODS

In partnership with Wayne State University, a simpliﬁed skull geometry was
meshed using the Hexmesher function in DEP Hexmorpher v4.0 (Detroit Engineered
Products, Detroit, Michigan) (Figure 1). The diameter of the simpliﬁed skull geometry
(80 millimeters) was based on the skull size of a 7 day old porcine specimen using CT
scans ﬂom specimens in Chapter 2. An element size of 0.5 mm was selected, yielding
36,525 solid elements. The outer faces of the hexahedral solids were duplicated to create

an overlying skull shell layer with nodal connectivity.

52

 

 

[IIIIIIIIILILILTLLLIJ

 

Figure 4.1. Simpliﬁed skull geometry used for ﬁnite element simulations. Rigid
impactor positioned above the skull.

Pre-processing was performed in Altair Hypermesh v9.0 (Altair Engineering,
Inc., Troy, Michigan). The solid elements were assigned viscoelastic properties
representative of brain tissue (p=l,040 kg/m3, K=211 MPa, (30:1.72 kPa, G1=0.51 kPa,
decay constant = 20 ms) (Mao et al., 2006). The bulk modulus has limited effect on
model predictions, but reducing it by a factor of 10 (ﬂom 2.1 GPa) saves computational
time (Kleiven, 2002). The brain was added in the model to include intracranial pressure
and displacement effects developed by the brain during a blunt impact. This method has
been used in previous studies to generate a better predictive model for head impacts
(Horgan and Gilchrist, 2004). The skull shell was assigned elastic properties of E=5 GPa
(Baumer et al., 2009), p=2,150 kg/m3, and u=0.28 (Margulies and Thibault, 2000) ﬂom
the literature. The total head mass was 359 g, which was the average mass of a 2-9 day
old pig head (see Appendix B). A skull thickness of 2 mm obtained ﬂom the CT scan
data was assigned to the shell layer. An 80x80 millimeter rigid impactor with aluminum

material properties (p=2700 kg/m3, E=70 GPa, u=0.35) was positioned 1.4 millimeters

53

above the sphere to allow initial translation before contact. The rigid impactor had a
mass of 1.67 kg, the same mass as the impactor in Chapter 2. Automatic surface-to-
surface contact in LS-DYNA (Livemore Software Technology Corporation, Livermore,
California) with no penalty stiffness was used to deﬁne interaction between the impact
interface and sphere. A simulation length of 4 milliseconds was set to show the entire
impact duration for both scenarios. The simulations were solved on an Opteron cluster
using 1 node (8 CPUs) in LS-DYNA 971r4. Post-processing software (LS PrePost 3.0,
Livermore Software Technology Corporation, Livermore, California) was used to assess
impact duration, maximum tensile and compressive stress magnitudes, and principal
stress directions for each simulation.

Two scenarios were simulated to compare the impact conditions in Chapter 2 and
3. In the ﬁrst simulation, half of the shell nodes were constrained in the simpliﬁed skull
to simulate the entrapped head. The constraint condition prevented the simpliﬁed skull
ﬂom translating during the impact. The rigid impactor was assigned an initial velocity of
2.8 m/s to represent a drop height of 40 centimeters. This drop height was used for 3-9
day old entrapped specimens in Chapter 2. In the second simulation, the rigid impactor
was constrained to represent an inﬁnite mass. The simpliﬁed skull displacement
constraints were removed and were given an initial velocity condition of 6.1 m/s to
represent a drop height of 1.87 meters. The drop heights were representative of equal
impact energies with respect to the mass of the impactor and the head and a static
gavitational acceleration (9.81 m/sz).

Fracture length, impact force, and impact duration data were revisited ﬂom

Chapters 2 and 3 for comparison. Effects of the impact scenario were assessed for

54

statistical signiﬁcance using statistics software (SignaStat 2.03, Aspire Software
International, Ashbum, VA). Two-way ANOVA tests were used to compare the
statistical data between the two scenarios. Statistically signiﬁcant effects were reported
for a p-value of less than 0.05.
RESULTS

Total ﬂacture length, impact duration, and peak impact force data ﬂom Chapters
2 and 3 were compared. There was signiﬁcantly more total skull ﬂacture at each age in
the entrapped head impacts than in the ﬂee fall experiments (p<0.001) (Figure 4.2).
Comparing the linear regession slopes and intercepts, the slopes were not found to be

signiﬁcantly different (p=0.2106), however, the intercepts were statistically different

 

 

 

(p<0.05).
_ H Entrapped D - -EI Free Fall
200
a ' '
é1eo- - I
5 I
U) I E l
5120- - ' ' . -
J “...-”M
e _ .
:r
'8 804 n 5 '
E n . '3 '
ll. 0 . . '
.‘3 401 "Ira-0 ----- a ........ 0 c1 0 3
O 0 O ....... D
I- n . I 8"'°Q'1E--a..,
El
0 a—C a H .
0 5 10 15 20
Age(Days)

Figure 4.2. Total ﬂacture (diastatic and bone) length with respect to specimen
age for both impact scenarios.

A two-way ANOVA analysis showed that the revisited peak impact force data

ﬂom Chapters 2 and 3 was not statistically different between the ﬂee fall and entrapped

impacts (Figure 4.3).

55

2500 _ H Entrapped E} - -Cl Free Fall

 

 

 

g 2000 J

0 I I

2

,2 1500 -

‘6

«I

@1000 -

3

to

g 500 -

D
O I I l I
0 5 10 1 5 20
Age (Days)
Figure 4.3. Peak impact force with respect to age for both ﬂee fall and entrapped
impacts.

Impact duration was deﬁned as the total time taken until the peak impact forces

began to decrease (Figure 4.4).

 

1000 '1 —Entrapped - Free Fall
Impact Durations
g 750 - N
0
g
u. 500 -
‘6
(U
Q
g 250 l ,
O I I I f I I I “ITAI'AMTTT'TT 1....

 

 

I

0 1 2 3 4 5 6 7 8 9 10 11
Time (milliseconds)

Figure 4.4. Force versus time plots used to determine the duration of impact for both
entrapped and ﬂee fall impact scenarios.

56

The impact duration was signiﬁcantly shorter in the experimental ﬂee fall

impacts for a given age than the experimental entrapped impacts (p<0.001) (Figure 4.5).

Linear regession analysis showed shorter impact durations with an increase in age for

botln scenarios. The theoretical impact durations taken ﬂom the ﬁnite element

simulations were 1.1 ms and 4 ms for the ﬂee fall and entrapped impacts, respectively.

 

I—I Entrapped D - 4:! Free Fall

 

8-
I
10‘ I
5.6- ' . ' .
I
g I
'54- 5 n :-
D
e D I 0 '
8 ........... r3 u I '
Q2- 0"D-n ..... E] """" Ifl""‘P--- B B D
E 0 I: D U .......
— U 0 :1 :1 B
D
0 D I I I I
O 5 10 15 20
Age(Days)

Figure 4.5. Impact duration with respect to age for the ﬂee fall and entrapped impacts.

The ﬁnite element simulations for the entrapped simpliﬁed geometry showed

maximum tensile stresses in areas remote of the impact site (Figure 4.6).

57

 

Figure 4.6. Entrapped simulation showing impact site (bull’s-eye) and surrounding
principal tensile stress directions. Four primary areas of maximum principal tensile
stress were documented. Darker arrows represent higher magnitudes of tensile stress.
These principal tensile directions and magnitudes were overlaid on a photo of an

entrapped porcine skull to compare the distribution of stresses to the experimental

ﬂactures (Figure 4.7).

 

Figure 4.7. Overlaid maximum tensile stress magnitudes and directions on a typical
pattern of ﬂactures ﬂom an experimental entrapped porcine specimen. Shaded areas
indicate documented ﬂacture with each line representing 5 millimeters.

58

For the ﬂee fall simulations, maximum principal stress magnitudes were located
nearer the point of impact (Figure 4.8). At points further ﬂom the point of impact, the

tensile stresses dimirnished rapidly.

I

 

I

Figure 4.8. Free fall simulation showing impact site (bull’s-eye) and surrounding
principal tensile stress directions. The largest magnitude of tensile stress was located
near the point of impact.
The ﬂee fall maximum principal stresses were also overlaid onto an

experimentally impacted ﬂee fall specimen to compare the observed patterns of ﬂacture

with the principal stress magnitudes and directions (Figure 4.9).

 

Figure 4.9. Overlaid maximum principal tensile stresses on an experimental ﬂee fall
porcine skull.

59

Maximum pressures of the brain were an order of magnitude higher in the
entrapped simulation than the ﬂee fall pressures (5.41 MPa in the entrapped versus 500
kPa in the ﬂee fall).

DISCUSSION

In this chapter, data ﬂom the previous two chapters was compared to assess the
differences between an entrapped head impacted with a mass and a ﬂeely falling head
impacting a rigid surface. The total ﬂacture length for the ﬂee fall impacts was
signiﬁcantly less than that for the entrapped impacts for an equal high level of impact
energy. The larger ﬂacture length in the entrapped scenario was attributed to ﬂacture in
the surrounding bones of the skull. In the ﬂee fall scenario, the ﬂactures were primarily
located in the parietal, although some ﬂacture occurred bilaterally in the ﬂontal bone
(Figure 3.7a). Peak impact forces were typically similar for a given age and impact
energy, and thus the increased amount of ﬂacture in the entrapped scenarios was not
attributed to a higher force of impact. Impact duration, however, was signiﬁcantly longer
in the entrapped specimens than in the ﬂee fall. Additionally, the maximum principal
stress distributions in the entrapped impacts encompassed a much larger area whereas
the ﬂee fall stresses were localized around the point of impact.

In the entrapped impacts, a mass was dropped onto a porcine head embedded in
epoxy to prevent translation of the head. These impacts were signiﬁcantly longer in
duration than the ﬂee fall impacts. In the literature, it has been shown that the density
and size of cracks in a brittle solid is affected by the magnitude and duration of an
applied dynamic elastic stress ﬁeld (Evans et al., 1978). Seaman et al. (1976) suggest

that propagation of ﬂacture is related to the stress and strain duration. Davis (2000)

60

documents that the duration and magnitude of the impact forces will determine the
extent of crarnial injury. Further, Gurdjian (1975) documents that higher and longer
intracrarnial pressure changes can increase the susceptibility of the skull to ﬂacture
(Figure 4.10). The intracranial pressures found in the entrapped ﬁnite element model
support this claim. The entrapped impact scenario produced ten times as much

intracranial pressure as the ﬂee fall.

 

COMPARISON OF FIXED AND FREE HEAD

 

DEGREE OF CONCUSSION

 

MIN MOD SEVERE

ks:

 

REMARKS

Higher and longer
duration intracranial
pressure changes
more susceptible to
fracture.

 

 

 

 

 

 

j.‘ j

    

/7

Figure 4.10. Gurdjian’s (1975) comparison of the degee of head trauma
resulting ﬂom a blunt impact to the head. The study suggests that longer duration
impacts increase susceptibility of the skull to ﬂacture.

These claims seem reasonable to apply to the current study where there are signiﬁcant
differences in experimental impact duration between two different impact scenarios
which generate dramatically different amounts of skull ﬂacture length. The time
durations recorded in the theoretical ﬁnite element simulations ﬁrrther imply that the
process of skull ﬂacture is a time-dependent phenomenon. Mechanically, a longer
impact duration places the skull under an elevated stress ﬁeld for an extended lengtln of

time. Assuming the bone ﬂactured at a ﬂaction of the applied stress ﬁeld, the initiation

ﬂacture would continue to propagate under the applied stress ﬁeld irn order to maximize

61

the energy dissipation. This may explain why there was more propagated ﬂacturing in
the entrapped high energy impacts than the ﬂee fall impacts where only ﬂacture
initiation seemed to have been documented. Further study of ﬂacture models is
necessary to verify these claims.

The maximum principal stress magnitudes and directions correlated with the
experimentally observed ﬂacture patterns, especially for the entrapped scenario. F rank
and Lawn (1967) claim that the geatest amount of energy dissipation occurs when the
crack plane is perpendicular to the maximum principal stress direction. They also
propose that ﬂacture is the key mechanic in maximizing energy dissipation. In the
entrapped scenario, the maximum tensile stresses were documented in areas surrounding
the impact site. When overlaid onto a porcine skull, the maximum stresses were not only
in the parietal bone but in the occipital bone as well. The principal stress directions were
also perpendicular to the key ﬂacture sites. In the ﬂee fall simulations, the maximum
tensile stresses were located primarily near the point of impact. Using the Frank and
Lawn approach, this would suggest that the skull ﬂactures should have been located near
the point of impact or in the immediate surrounding area. However, the ﬂee fall impacts
typically generated small ﬂacture initiation sites away ﬂom the point of impact near the
sutures surrounding the parietal bone. The mechanism of ﬂacture is therefore somewhat
uncertain. The simulation geometry is restricted in that it is a homogeneous bone layer
with a brain solid interior. There were no sutures with appropriate material properties to
affect the principal stress & strain distributions across the skull (Herring and Teng,

2000). It may be possible that the ﬂacture initiation occurs at the bone-suture boundaries

62

due to stress and strain concentrations in the bone-suture interface (Baumer et al., in
press; Yu et al., 2004; Alexandridis et al., 1985).

In summary, the ﬂacture length, impact duration and peak impact force data ﬂom
the previous chapters was compared. Fracture length was much geater in the entrapped
impacts than the ﬂee falls. This was attributed to the entrapped impacts having longer
impact durations. Longer impact durations promote ﬂacture propagation in an attempt to
maximize the energy dissipation in the system. Also, a ﬁnite element software package
was used to simulate the two impact scenarios using a simpliﬁed skull geometry to
validate the increased amount of ﬂacture length documented in the experimental
entrapped skull impacts. It was found that an entrapped impact produced a larger
distribution of maximum principal tensile stresses across the entire skull, whereas the
ﬂee fall impacts generated a more localized maximum principal stress pattern near the
point of impact. The model, however, was limited in geometry and lacked interfacial
changes between the bone and sutures. In order to better understand the mechanisms of
remote ﬂactures near the sutures in the experimentally observed ﬂee fall scenarios, a
more complex model of the immature skull is likely required. Furthermore, the current
study suggests that mechanisms of bone and suture damage in the pediatric skull may be
very dependent on temporal stress-strain effects. A purely elastic analysis may therefore

not be appropriate.

63

REFERENCES

Alexandridis C, Caputo AA, Thanos CE, 1985, “Distribution of stresses in the human
skull,” Journal of Oral Rehabilitation, 12(6), pp. 499-507.

Baumer TG, 2009, “Material Property Documentation and Fracture Analyses of the
Developing Skull,” Masters Thesis Dissertation. East Lansing (MI): Michigan State
University.

Baumer TG, Nashelsky M, Hurst CV, Passalacqua NV, Fenton TW, Haut RC, in press,
“Characteristics and Prediction of Cranial Crush Injuries in Children,” Journal of
Forensic Science.

Baumer TG, Powell BJ, Fenton TW, Haut RC, 2009, “Age Dependent Mechanical
Properties of the Infant Porcine Parietal Bone and a Correlation to the Human,”
Journal of Biomechanical Engineering, 131(11), pp. 1-6.

Davis AE, 2000, “Mechanisms of Traumatic Brain Injury: Biomecharnical, Structural and
Cellular Considerations,” Critical Care Nursing Quarterly, 23(3), pp. 1-13.

Doorly MC and Gilchrist MD, 2006, “The use of accident reconstruction for the analysis
of traumatic brain injury due to head impacts arising ﬂom falls,” Computer Metlnods
in Biomechanics and Biomedical Engineering, 9(6), pp. 371-377.

Evans AG, Gulden ME, Rosenblatt M, 1978, “Impact damage in brittle materials in the
elastic-plastic response regime,” Proceedings of the Royal Society of London, A361,
pp. 343-365.

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

Gurdjian ES, 1975, “Impact Head Injury,” Charles C. Thomas, Springﬁeld, IL.

Hall JR, Reyes HM, Horvat M, Meller JL, Stein R, 1989, “The Mortality of Childhood
Falls,” The Journal of Trauma, 29(9), pp. 1273-1275.

Herring SW and Teng S, 2000, “Strain in the braincase and its sutures during function,”
American Journal of Physical Anthropology, 112(4), pp. 575-593.

Horgan TJ and Gilchrist MD, 2004, “Inﬂuence of FE model variability in predicting
brain motion and intracrarnial pressure changes in head impact simulations,”

International Journal of Crashworthiness, 9(4), pp. 401-418.

Kim KA, Wang MY, Grifﬁth PM, Summers S, Levy ML, 2000, “Analysis of pediatric
head injury ﬂom falls,” Neurosurgical Focus, 8, Article 3.

64

Kleiven S, 2002, “Finite Element Modeling of the Human Head,” Doctoral Thesis
Dissertation. Stockholm, Sweden: Royal Institute of Technology.

Mao H, Zhang L, Yang K, King A, 2006, “Application of a ﬁnite element model of the
brain to study traumatic brain injury mecharnisms in the rat,” Stapp Car Crash
Journal, 50, pp.583-600.

Margulies S and Thibault K, 2000, “Infant Skull and Suture Properties: Measurements
and Implications for Mechanisms of Pediatric Brain Injury,” Journal of
Biomechanical Engineering, 122(4), pp. 364-371.

Seaman L, Curran DR, Shockey DA, 1976, “Computational models for ductile and
brittle ﬂacture,” Journal of Applied Physics, 47(11), pp. 4814-4826.

Silva MJ, Keaveny TM, Hayes WC, 1998, “Computed Tomogaphy-Based Firnite
Element Analysis Predicts Failure Loads and Fracture Patterns for Vertebral
Sections,” Journal of Orthopaedic Research, 16(3), pp. 300-8.

Yu JC, Borke JL, Zhang G, 2004, “Brief synopsis of cranial sutures: Optimization by
adaptation,” Seminars in Pediatric Neurology, 11(4), pp. 249-255.

65

CHAPTER FIVE

CONCLUSIONS AND RECOMMENDATIONS FOR FUTURE WORK

This thesis documented the ﬂacture characteristics of the infant porcine skull as a
function of specimen age, impact interface, level of impact energy, and impact scenario.
The ﬂacture patterns resulting ﬂom a high energy blunt impact to the parietal bone were
documented using GIS software. Porcine heads were dropped ﬂom a controlled height
onto a rigid interface to assess the differences in the pattern of skull ﬂacture between
two impact scenarios: ﬂee fall and entrapped head impacts. Finally, a simpliﬁed skull
geometry was modeled in a ﬁnite element software package to investigate potential
explanations for differences in the ﬂacture lengths and distributions of ﬂacture
documented in the entrapped and ﬂee fall head experiments.

In Chapter 2, high energy blunt impacts of infant porcine skulls were conducted
in order to study characteristic patterns of propagated skull ﬂacture. It was found that
new ﬂacture initiation sites were generated in these higher energy impacts that were
dependent on impact interface. A rigid interface produced numerous new ﬂacture
irnitiation sites, while the compliant interface produced only one. Additionally, the rigid
interface produced more skull ﬂacture than the compliant at all ages except for 2 days
for an equal amount of impact energy. GIS was used to document the ﬂequency of
characteristic ﬂactures sites generated for both interfaces and at high and low (revisited
data ﬂom Baumer et al, in press) impact energy levels. Several urnique characteristic
ﬂactures were documented for each interface, age and energy level impact. One example
of these characteristic ﬂactures was a rigid mass striking a constrained head with a high

level of impact energy produced a pattern of skull ﬂacture including parietal, occipital

66

and diastatic ﬂactures. This combination and location of ﬂacturing was unique to only
the high energy, rigid impacted specimens and was tlnus a characteristic ﬂacture pattern
for those impact conditions. If given a similar pattern in a case of potential child abuse, it
may help in diagnosing whether the trauma was accidental or inﬂicted. Further studies
should focus on continued development of these characteristic ﬂactures using different
ages, interfaces and impact energy levels. A comparative database describing ﬂacture
characteristics based on the input conditions may be used to eliminate or validate causes
of trauma in cases of potential abuse.

In Chapter 3, porcine heads were dropped onto a rigid mass using a gavity
accelerated drop trolley. It was found that the pattern of ﬂacture was different than that
of the entrapped head impacts for an equal level of impact energy. Fractures in the ﬂee
fall head drops were typically located in the parietal bone, however in younger aged
specimens the ﬂactures extended bilaterally into the ﬂontal bone, producing ﬂacture on
either side of the coronal suture (Figure 3.7a). The suture is thought to transmit the
impact stresses to surrounding bones by its inherent stiffening during high rates of
loading which may help explain the bilateral ﬂactures shown in the younger specimens.
In the older age goup, the sutures are more developed and the impact stresses are likely
distributed more uniformly across the entire skull, reducing the potential for skull
ﬂacture. Extensive damage at the coronal suture was documented regardless of age so it
is reasonable that a large portion of the impact stresses were located near or were
transmitted through the coronal suture. Occipital ﬂacture was not documented in the ﬂee
fall impacts which signiﬁcantly contrasted the high energy entrapped impact ﬂacture

patterns. These results suggest that the anterior parietal bone and coronal suture

67

boundary is the weakest section of the skull in tlnese impacts (Macdonald et al., 1988;
Burstein and Frankel, 1971). Ultimately, the differences in ﬂacture patterns documented
in this chapter could help in future diagnoses of whether a victim’s head was constrained
during impact. Future work should investigate the differences in skull ﬂacture patterns
due to different impact interfaces, impact energies, and impact locations using the ﬂee
fall porcine head model. A different impact energy or impact interface may produce a
different pattern of skull ﬂacture than documented in this study for a high energy, rigid,
ﬂee fall impact. A larger range of ﬂacture patterrns may produce characteristic features
of skull ﬂacture depending on each set of input conditions. These characteristic features
may prove to be extremely valuable for forensic pathologists and medical exanniners to
compare to potential child abuse cases to better diagnose the causation of trauma.

In Chapter 4, a simpliﬁed skull geometry was modeled using a ﬁnite software
package. Two simulations were conducted to assess the differences in maximum
principal tensile magnitude and directions between the impact scenarios presented in
Chapters 2 and 3. Total ﬂacture length, peak impact force, and impact duration data
ﬂom Chapters 2 and 3 were also compared. Total ﬂacture length was found to be
signiﬁcantly geater in the entrapped impacts than in the ﬂee fall. However, there was
little difference in peak impact force for a given age, and thus the increased ﬂacturing
was not attributed to impact force. The impact duration was signiﬁcantly geater in the
entrapped scenarios tlnen the ﬂee falls for a given age. The literature suggests that a
longer impact duration under a dynamic stress ﬁeld makes the skull more susceptible to
ﬂacture. Also, the maximum principal stresses were distributed over a larger area of the

entrapped simpliﬁed skull in the ﬁnite element simulations than the ﬂeely falling skull.

68

Using an overlay of these principal stress magnitudes and directions, it was noted that
the experimental skull ﬂactures compared well with the theoretical stress directions for
the entrapped scenario, however the ﬂee fall experimental ﬂactures did not resemble the
theoretical predictions. Further studies should use a more complex and realistic skull
geometry taken ﬂom a computed-tomogaphy (CT) scan to investigate principal stress
differences due to the more complex geometry. The model should also include sutures to
assess how the impact stresses are affected by a non-homogeneous material and interface
conditions between the bone and suture. It has been shown that increased levels of strain
exist at the sutures during impact which may be a key indicator of why ﬂacture initiation
occurs at the bone-suture boundary in the ﬂee fall impacts (Baumer, 2009).

This research was based on assessing the degee and pattern of ﬂacture of an
infant porcine head model impacted in two injury scenarios. It was found that specimen
age as well as impact interface, energy, and scenario all contribute to the pattern and
degee of skull ﬂacture in a porcine head model. This work suggests that investigating
characteristic patterns of ﬂacturc may aid in producing a more robust predictive ﬂacture
model to help diagnose accidental trauma ﬂom inﬂicted. Pattern recognition techniques
currently used for ﬁngerprints could potentially help quantify these characteristic
ﬂacture patterns for each set of impact conditions. With a database of characteristic
ﬂacture patterns for a wide range of impact conditions, a forensic pathologist or medical
examiner may be able to ﬁnd a causation of trauma based on a victim’s skull ﬂacture

pattern.

69

REFERENCES

Baumer TG, 2009, “Material Property Documentation and Fracture Analyses of the
Developing Skull,” Masters Thesis Dissertation. East Lansing (MI): Michigan State
University.

Baumer TG, Passalacqua NV, Powell BJ, Newberry WN, Fenton TW, Haut RC, in
press, “Age-Dependent Fracture Characteristics of Rigid and Compliant Surface
Impacts on the Infant Skull — A Porcine Model,” Journal of Forensic Science.

Burstein AH and Frankel VH, 1971, “A standard test for laboratory animal bone,”
Journal of Biomechanics, 4(2), pp. 155-158.

Macdonald W, Skirving AP, Scull ER, 1988, “A device for producing experimental
ﬂactures,” Acta ortlnopaedica Scandinavica, 59(5), pp. 542-544.

70

APPENDIX A
RAW DATA FROM CHAPTER 2

71

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

72

o F .8N ova _ Nmood ON. 3 _ no.2: 83 ON 3 N rowIE
Snow 23 _ 28.0 o; _ 38$ 8.2 E 3 38:3
863 2.0m _ mvood om: _ 8.3? map on 3 voomIE
chmEmn :36 9 can corona. EB 68:: - 02:89 6: See .cOnocEﬁE BSanO mom? on 3 881.1
deN vmvd :86 E: New: 0 5? co me 805
2.8? mmmé 886 SN? vaNo 33 oo 2 82¢
omdNN mmmd Nmood one”: mode moor 0N NF 9 Ed
woNNF 086 9.8.0 2.3 omdmo 89 on Z on w E
EN: 3.2m mmood mva ovNE moor ov m 9.815
9.va NmN.m Rood 0N.m oodom mom: ov m 381?.
8.09 Nmné Nmood mvNF NN. F 5 moor ov m N F _n_
5.2 ONm.m $86 NmNF :68 wow? ov w or :a
$69 mwod $86 3.2. 3.39. we? O... m Somzi
wad? NONB mvood mm. 3 omhnw moor ov w meow—... a
5.8? meme 9.8.0 no.2. wo.mmm moor ov w Noomri
IcmmEme :36 2 can corona. :5 SEE - noncooo. Co: San 60:95sz .93 E00 meme ov w 581?.
mNm. 5v momN 586 or .m N F .moo mom? 9... N 881?.
mu. 5N Now.”c 9.8.0 vaF 56o? moor ow N 8315
ENNN 25m 88.0 8.3 moeNNF moor ov N moomzi
vo. 5F vONd Nvood NN.o_. mN.mmo moor ov N 2. :a
tee mmmb mvood me. 5 2.33 mom: ov m N Four-n.
$.va Secs ovood Ede mmNNm moor ov m mN:n_
8.3; mmud Nvood 8d mNdmN moor ov m mNza
NN.m_.N mine mmood mNdF 5. SK moor ov v m—omIE
mon {mm 3.8.0 8.3 FNdNN moor ov m 82d
No.0: NmNN mmood Rd NNdmv vow? 0N N 89d
.53
:5sz 3 3 Ass. 80.. .3: A2. neon a. news. 292.. was om< 558nm
mmoctzw >925 95.3.... c2359 BEE. “a EoEoermE Seas: mcEen. no.0

 

 

 

 

 

 

 

 

 

 

 

.3895 BEBE: Emu Q85 :mE 88m See 3mm .~.< «Ssh.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

73

3.3m moms 386 mm. 3 mo. r EN m5? 8.. N 531:.
NméFN 5% m Sod Noe? 5.82. 32. ON? mN omomIE
mm. 5m Se. 3 $86 8.3. $.89 33 on N NNomIE
3.6mm. can.» 38.0 mo. 3 3.3.: a SF on N oNomIE
mNémF N _.m.m 2.86 8.2. Sam FNF m ..mp co ON mNomIE
Iodemc =3...“ 0. can Stone. :5 “08:: - 3280.. Doc Emu 60:05sz .65an0 9.3 cm oN NNomI_n_
mN. FNN vmod 58.0 «6.9 N tmmNF 2.3 oo 3 38:5
VNNt. Vde oeood em. 2. madam a z: co 3 32a
endoN ONNé Need 96 mmNNm meme om 5 BBQ
3N2. Ems 2.8.0 NN.3 v0.89 mom? on or 98.13
:63 ommd $.86 mm.N_. 8.8m mom? on or E :a
m... :.m N 56 $86 3.5 mN.vmmN mom? on or m _.om_I_n_
A53
...:sz S 3 Es. .23.. ca: .2. 83a. E 8e: E 22.. in... eu< 55.8%
32.55 >925 95.3“. coach—5 «can—E. «a EoEoanwS «owns: 9.5a". no.5

 

 

 

 

 

 

 

 

 

 

 

was—5.80 .m.< 035—.

Table A.2. Thickness, ﬂacture length, and contact area measurements for high energy

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

rigid interface impacts.
Age Average Bone Suture Total Contact

Specimen (da ) Thickness Fracture Damage Damage Area
ys (mm) (mm) (mm) (mm) (mm’)

P1058 2 1.28 72 18 90 64.2
P1109 3 1.03 118 16 134 264.9
PIHE018 4 1.05 135 25 160 463.1

P1125 5 1.15 39 20 59 41.0
P1129 5 1.18 120 0 120 155.3
PIHE017 5 1.14 50 0 50 326.0

P1113 7 0.99 80 45 125 53.7
PIHE006 7 1.33 140 0 140 443.3
PIHE008 7 1.15 65 0 65 256.2
PIHE009 7 1.12 135 0 135 478.8
P1HE001 8 - 85 20 105 207.3
PIHE002 8 - 73 7 80 146.3
PIHE005 8 1.47 12 0 12 276.6
P1HE007 8 1 .43 45 0 45 451 .8
P1110 8 1.06 90 55 145 225.6
P1126 9 1.42 15 0 15 103.6
PIHE015 9 1.99 95 0 95 339.0
PIHE016 9 1.27 130 0 130 251.5
P1130 1 1 0.95 30 25 55 NO PF
P1116 12 1.64 123 0 123 349.2
P1002 13 1.84 95 21 116 158.6
P1006 13 1.02 190 0 190 120.4
PIHE003 14 - 80 22 102 224.0
PIHE004 14 1.73 80 0 80 155.3
PIHE011 15 2.00 155 0 155 622.9
PIHE012 15 1.93 20 0 20 492.0
PIHE013 15 1.08 115 37 152 235.6
P1111 16 2.01 95 27 122 278.8
PIHE010 16 2.28 185 0 185 549.1
P1055 17 1.53 40 0 40 248.6

P1070 19 1.41 220 0 220 61.9
PIHE014 19 1.78 35 0 35 520.3
PIHE022 20 2.28 105 0 105 580.8
PIHE023 20 2.73 225 15 240 695.2
PIHE026 24 2.91 90 0 90 628.2
PIHE027 24 2.72 165 0 165 706.9
P1HE030 28 2.59 65 0 65 584.9
PIHE031 28 3.92 55 0 55 810.1

 

 

 

 

 

 

 

 

 

74

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

vde 08.3 88.0 ..wNN _ 5.39. 32. ON? N 381?.
me. For Nmmd 030.0 00.3 _ 3462 m 5.. oNF wN NmomIE
2.1m? $93 38.0 8.: _ 3.32 33 ON N 38.15
.Iemm Emu :36 0. can Stone. :3 .895 - 80.80. 8.. Eco 628..ng .33 E00 32. oN VN mNomIE
$69. Na. 3 $86 8.2. New: m 5.. ON VN wNomI_ d
oN.mNN owmd 9.8.0 vNNv 86wa 39 cm 0N mNomIE
8N? mNNd Nmood 3.9 Nm.wNN_. 32. co 0N «NomIE
mNde on: 2.86 86 Eva: m5? co 2. veomIE
N53: ommd 9.8.0 8.3 m Sb: 89. ON me 882..
Ned? Ne mm 2.86 8.9 5. Km 82. ON ..F Nmom; a
ohmw ommd mvood w _..m_. 0N. _.oo_. 32. cm or meow T: a
1emmEmu :33 9 can Stone. :5 Same: - 8282 So: Emu 62.2.59: .23 :80 wow? cm or FNomIE
56m comm 38.0 mm: 5N8 mom: 2. m 3815
NmN: NNN.m 38.0 3.3 3.3.0 moor ow N NvomIE
m 3m Name mNood N v.3 no. 5N moor ov o mmom I_ n.
mm. NF 08.. N Sod 86 woNoN moor o... m 381E
NN.mw momé 88.0 no.3 wodmm meme 9. m 88.15
8.8? 056 wNood N3; odeN 83 o... v 2.815
863 N94. 9.86 3.9. oodmN 89. ow m m Pom—IE
:5.
2:25 3 3 2:5 28.. as: .2. 83. a. can: 292.. .233 oa< 55.8%
«caesium >925 22.5.. c2350 32:... «a 22.53320 82:... 9...?“— .35

 

 

.Soemﬁm coerce: Sewage—.8 3.28 .33 Bob 8% 26M .m.< 033-.

75

Table A.4. Thickness, fracture length, and contact area measurements for high energy

compliant interface impacts.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Age Average Bone Suture Total Contact
Specimen (da ) Thickness Fracture Damage Damage Area

’3 (mm) (mm) (mm) (mm) (mm’)
PIHE019 3 1.18 90 10 100 756.2
PIHE040 4 1 .46 45 20 65 464.5
PIHE020 5 1 .08 55 15 70 796.3
PIHE041 5 1.23 95 20 115 842.7
PIHE039 6 1.31 30 25 55 472.9
PIHE042 7 1 .02 105 15 120 724.3
PIHE038 8 1.64 45 0 45 621 .9
PIHE021 10 1.66 175 0 175 480.3
PIHE043 10 1.58 100 0 100 906.1
PIHE037 11 2.42 5 0 5 844.0
PIHE036 13 2.31 10 0 10 580.5
PIHE044 18 1.38 145 25 170 995.2
PIHE024 20 2.23 100 0 100 824.6
PIHE025 20 2.37 35 0 35 918.9
PIHE028 24 2.18 35 0 35 687.5
PIHE029 24 2.27 115 0 115 807.2
PIHE034 24 2.51 0 0 0 1133.1
PIHE032 28 2.83 70 5 75 1311.6
PIHE033 28 2.62 0 0 0 1394.9

 

76

 

APPENDIX B
RAW DATA FROM CHAPTER 3

77

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

888 88.8 88.8 8.8 8.88 88 8.82 3 3888
8.88 «88 8 188 8.8 8.83 88 8.88 3 3888
8.82 88.8 88.8 8. : 8.88 88 8 : 8 m 88 S
8.02 88.8 888 8.8 2.88 88 8.8: m? 28.88
2.88 888 S88 8.8 8.8: 88 3.2 8 88. 8
8. SF 888 888 8.8 8.88 88 8.3: 8 8858
88: 82. :88 8.8 8.88 88 N82 : $888
8.3: 82.8 8 88 8.8 8.88 88 8.82 : $8.88
:8: 88.8 :88 8.2 :88 88 8.82 2 28:8
8.8 88.8 888 8.: :88 88 8.82 S 8 E88
8.88 88.8 288 8.8 8.88 88 8.8: m 88.88
8.8 88.8 288 8.2 2.88 88 8.8: m 8888
8.88 88.8 28.8 8.8 8.88 88 83 8 888.8
888 88.8 888 8.8 8.8 E 88 39 m 88: 8
8.8 83 888 8.3 8.88 88 8.8? 8 88:8
8.8: 888 8.8.8 8.8 8.8: 88 8.88 8 8888
8.8: ~88 88.8 8.8 88 E 88 :2 o 888 8
8.8. 888 288 8.8 8.88 88 :2 8 88:8
8.8 88 88.8 8. : .8. 8 88 8.82 8 8888
8.82 88.8 888 8.2 8. 8 88 8.89 m 88.88
8.8 8: 88.8 8.8 8.8 8 88 m. 8 m 88_ 8
8.88 88 88.8 8.8 8.88 88 we: 8 88.88
888 88 88.8 8. 3 8.88 88 N8: 8 8888
8.88 88.8 :88 8.8 8.88 88 3.: m «888
88 F F 88.8 888 8.8 8E 88 8.8: m 88. 8
8.82 88.8 888 8.8 8.88 88 8.8 N 88.88
.82.: S 3 8:5 .83 .8: 22. 88". a. a8: .53 883
222.. 58325
30525 >Eocw 95:3. acacia «03...: «a EeEeowEeE 33E. new... no.5 o2

 

 

 

 

 

 

 

 

 

 

.888 8888 28 :8 88 88 88:8 88 38 .3 28.8

78

 

 

 

 

 

 

 

m ... BF 88.0 omood om. Pm "odour mac 8. EL 5 n {ammo
coda «82¢ mmood 5.8 8.3:. on» 5va or 0 Kim 3
mad? Rad ..wood 56w 8.03: on» ~89 S. 0 Kim 3
wmdmv 89m wooed o P. S. 8.89 com 88: 3 m FEV 8
mod? «mod owood mm. E 3.38. com 9va 2 m Fin 5
Eu
.22... .3 ... .52. .23.. z... .z. 83.. a. .3: . . E3.
222.. :25“.on
23525 .535 95:3. 20.3.59 «03:: «a 22.33.33 «can»... .52.. no.5 02

 

 

 

 

 

 

 

 

 

 

62.5880 .—.m ﬂank.

79

Table 8.2. Fracture length and contact area measurements for free fall high energy rigid

interface impacts.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Age Bone Suture Total Contact
Specimen (da ) Fracture Damage Damage Area
’3 (mm) (mm) (mm) (m_m’)
OO3FF02 2 10 0 10 13.828
021 FF03 3 40 20 60 44.004
022FF03 3 17 15 32 190.561
024FF04 4 32 0 32 61 .469
025FF04 4 52 0 52 47.698
001 FF05 5 5 30 35 197.919
002FF05 . 5 65 10 75 58.892
005FF05 5 0 0 0 61 .240
017FF06 6 0 0 0 144.123
018FF06 6 55 10 65 129.408
026FF07 7 20 10 30 30.863
027FF07 7 15 20 35 98.058
011FF08 8 30 45 75 91.559
012FF08 8 10 15 25 167.027
004FF09 9 15 0 15 52.708
008FF09 9 15 0 15 153.543
006FF10 10 6 20 26 66.049
007FF10 10 0 0 0 144.925
028FF11 11 5 0 5 129.064
029FF11 11 30 0 30 188.786
030FF12 12 15 0 15 71.489
031FF12 12 10 0 10 77.874
009FF13 13 5 25 30 183.146
010FF13 13 15 10 25 112.144
019FF14 14 20 0 20 328.214
020FF14 14 17 22 39 260.619
013FF15 15 15 0 15 253.061
014FF15 15 0 0 0 205.134
015FF16 16 0 0 0 238.860
016FF16 16 15 0 15 133.072
023FF17 17 30 0 30 140.602

 

80

 

MICHIGAN STATE UNIVERSITY LIBRARIES

ILIIIIIIIH

3 1293 (13 63 6800