Content uploaded by Jennifer A Himmelsbach
Author content
All content in this area was uploaded by Jennifer A Himmelsbach on Jun 20, 2014
Content may be subject to copyright.
49
Sports surfaces and the risk of
traumatic brain injury
Martyn R. Shorten, PhD and Jennifer A. Himmelsbach, MS
BioMechanica, LLC, Portland, Oregon, USA
Introduction
Sports provide many opportunities for an athlete’s head to
experience an impact or other violent acceleration. Player to
player contact, a collision with a goal post or other structure or
striking the surface during a fall can all result in brain trauma.
Sports are a common venue for concussion, more formally
described as Mild Traumatic Brain Injury or MTBI. Indeed, the
US Centers for Disease Control and Prevention considers the
incidence of sports-related MTBI to have reached epidemic
proportions (CDC, 1977). Also, in many sporting contexts, the
threat of more severe, life-threatening head trauma is always
present.
While severe head injuries are relatively rare, they have the
potential to change lives in a dramatically negative way and
carry a greater risk of fatality than more common injuries.
Consequently, they have been a focus of attention in the
sports medicine community for many years. Concern about
more minor head injuries has also increased in recent years,
with the realization that MTBIs can have cumulative, long-
lasting effects on cognitive function and can also expose an
athlete to a period of greater risk of severe injury or death from
a second or subsequent episode.
The potential for head injury has an influence on the
development and marketing of sports surfaces, too. In several
cases, the standard test methods used to evaluate and specify
the shock attenuation of sports surfaces, crash mats and wall
padding are based on the assumption that the there is a risk of
head injury and that an appropriately cushioned surface can
reduce that risk. For example, a commonly used specification
for the shock attenuation of North American football fields
(ASTM F1936) is hypothetically linked to the non-fatal
50
acceleration tolerance of the head. Frequently, F1936 is the
only performance-related specification included in a field
purchase contract, suggesting that head injury risk is an
important consideration in the purchasing decision.
This paper briefly summarizes some of the medical and
research literature related to traumatic brain injuries in the
context of sports. We also attempt to determine the extent to
which sports surfaces represent a risk factor by reevaluating
the relationship between the outcomes of common surface
impact test methods and head injury risk.
Mechanisms of brain trauma
The extensive medical and biomechanical literature relating to
brain trauma in sports has been reviewed recently by Bailes
and Cantu (2001) and Wojtys et al, (1999). The fundamental
cause of most brain injuries is strain (compression or
stretching) of the brain tissue and the blood vessels it contains.
Historically, head injuries have been classified in various ways,
based on their severity and on the mechanical factors involved
in their aetiology. In general, brain injuries can be considered
as either “focal” or “diffuse”. Focal injuries are typically
confined to a local region of the brain and are usually the result
of a direct blow to the head. Even if it does not fracture the
skull, an event of this type produces a shock wave that
alternately compresses and stretches brain tissue, causing
local tearing of brain tissue and blood vessels. The
subsequent hemorrhaging and haematoma can be fatal.
While a direct blow to a stationary head will normally produce
an injury in its immediate vicinity, a focal injury does not
necessarily occur at the point of impact. If the head is moving
when it experiences a collision, the brain moves around inside
the cranium, where it floats in cerebrospinal fluid. Commonly,
the brain is traumatized by impact with the skull at a point
opposite the point of impact (a “contrecoup” injury). Motion of
the brain inside the skull can also lead to cavitation, the
formation of a vacuum that can disrupt brain tissue and small
blood vessels.
51
The brain trauma associated with a diffuse injury is often less
severe but usually more widespread than with a focal injury. A
direct impact can cause a diffuse brain injury but it is not
required. A collision in which the head is not directly involved,
a hard tackle or vigorous shaking for example, can cause the
head to rotate violently. The brain accelerates inside the skull
generating injurious strain levels in the process. (The Latin
verb meaning “to shake violently” is “concussus”, from which
the English word “concussion” is derived.) Mild diffuse injuries
such as MTBI are characterized by short term cognitive
dysfunction and possibly loss of consciousness. More severe
diffuse injuries can result in deep, permanent coma and have a
mortality rate in excess of 50%.
The relative importance of linear, translational motion and
rotational motion of the head in brain trauma mechanisms has
been the subject of some debate. It is possible that the
distinction between linear and rotational motion is arbitrary,
since real events involve some degree of both. A translational
acceleration of the brain is more likely to occur when there is a
direct impact. A rapid rotational acceleration is more likely to
result from an indirect blow and produce a more diffuse injury.
High rotational accelerations of the head also cause tearing of
the nerve cells in the region of the cerebrospinal junction. An
impact between the head and a sports surface can be
expected to result in accelerations that are primarily linear, but
rotational accelerations are also possible, depending on the
geometry of the head, neck and torso at the instant of impact,
and the friction of the surface.
The consequences of brain trauma sometimes emerge over
time. In the hours and days following the initial traumatic event,
physiological changes occur that can have far reaching
consequences. When brain cells are torn, calcium and
potassium ions escape into the surrounding interstitial fluid
(Katayama et al, 1990). Since nerve impulses are transmitted
by the flow of these ions across cell membranes, the released
ions can disrupt neural function. What follows is a “metabolic
cascade” of events as healthy cells try to compensate for the
52
uncontrolled flow of ions, demanding more energy and
consuming more glucose in the process. This metabolic
distress leaves the brain vulnerable for some time after the
initial injury, increasing the probability of a further injury.
Guskiewcz et al (2000), for example, found that football
players who had experienced a concussion were three times
more likely to experience a second concussion in the same
season.
Surfaces as a risk factor
The extent to which sports surfaces are a factor in brain injury
is unclear. The incidence of injuries of all kinds has been well
documented for most major sports and recreational activities,
but these studies rarely distinguish impacts with the surface
from other impacts, nor do they document the surface type or
condition involved in a head injury. However, it is reasonable
to believe that a collision between the head and a surface has
the same injury potential as a direct impact with any other
object and the limited information available supports this
assumption.
Falls to the surface account for 21% of the deaths in
playground equipment-related accidents and most of these
(~75%) involve catastrophic head injury (Tinsworth et al,
2001). “Unsuitable surfacing” has been found to account for
between 79% and 100% of severe head injuries (Mack et al,
2000).
It can also be shown that different surfaces present different risks
of head injury. For example, the risk of serious head injury
following a fall is 1.7 times greater on a grass surface than it is
on sand
(Laforest et al, 2000). Clarke et al (1978) found no
difference in the incidence of MTBI between natural and artificial
turf while Naunheim et al (2002) suggest that risk is higher on
artificial turf. Neither study presents convincing evidence,
however. More persuasive is the study of Guskeiwicz et al
(2000) who tracked injury rates among 17549 high school and
collegiate football players. They documented 1003 cases of
MTBI, of which 10% were due to impact between the head and
53
the playing surface. The rate of surface-related head injury per
1000 athlete-exposures on artificial turf was approximately
double that on natural turf. More significantly, 22% of the
concussive impacts on artificial turf resulted in Grade II injuries
involving loss of consciousness, compared with 9% of the
impacts on natural turf. This finding equates to a five times
greater risk of the more severe, Grade II MTBI on artificial turf.
Since both “natural” and “artificial” turf encompass a wide range
of surface properties the particular characteristics that caused
the difference in head injury incidence remains unknown.
Assessing head injury risk
The published research shows that impact is strongly
implicated in the etiology of traumatic head injury, that sports
surfaces present an opportunity for impacts to occur and that
different kinds of surfaces present different relative risks of
injury. Therefore, it is important to assess how different surface
designs and material properties can influence head injury risk.
Epidemiological studies that track sports injuries and
document the surfaces on which they occur would be very
helpful in this regard, but few exist. If they did, they would tell
us about existing, installed surfaces but would not provide a
means of evaluating new or prototype surfaces. Laboratory
studies with human subjects are also of limited value in this
context because the researcher has an ethical responsibility
not to expose subjects to the possibility of an injury. Under
these circumstances, it is a normal for scientists to use a
surrogate, instead of human subjects. Human surrogates
commonly used in head impact research include cadavers,
anesthetized animals, physical models (e.g. headforms or
crash dummies), mathematical models and computer
simulations.
Impact Tests as Human Surrogates
An impact test is an example of a physical model, and one that
is commonly used to evaluate the shock attenuation
performance of both sports surfaces and the protective
54
equipment used by athletes. An impact may be loosely defined
as a brief period of intense acceleration, such as may be
caused by a collision. A test of surface shock attenuation
simulates an impact by dropping an instrumented weight onto
the surface and measuring the resulting acceleration. The
acceleration is usually expressed in g’s, where one g is
equivalent to the acceleration due to gravity. One way to
quantify the magnitude of an impact is to measure the peak
acceleration it produces. This peak acceleration is commonly
referred to as the g
max
score (Figure 1A).
In order for an impact test to be useful in assessing the
potential risk of head injury, three requirements must be met:
1. The tolerance of the brain to impact loads must be
documented so that the relationship between impact
dynamics and injury risk can be quantified.
2. The impact test must simulate the potentially injurious
events that athletes might be exposed to during play. It
may be necessary to devise different tests for different
sports if they athletes are likely to experience different
collision dynamics.
3. There must exist a means of comparing the outcomes of
the impact test with impact tolerance data in a way that
produces meaningful information about surface
performance.
These requirements will be considered in more detail in the
following sections.
Impact tolerance of the brain
Early experiments on the ability of the human brain to
withstand impact were performed at Wayne State University
using human cadavers and animal models (Gurdjian et al,
1945, Gurdjian et al, 1955). This pioneering work eventually
led to the publication of the “Wayne State Tolerance Curve”
(Lissner et al, 1960; Patrick et al, 1963), a roughly logarithmic
curve that describes the relationship between the magnitude
and duration of impact acceleration and the onset of skull
fractures. The relationship is nonlinear – the head can tolerate
55
high accelerations for very brief periods but a longer exposure
to a lower acceleration level may be damaging. For a given
degree of injury the logarithmic slope of the exposure time /
acceleration graph is approximately –2.5. Gadd (1966) both
discovered and exploited this relationship, proposing the
Severity Index (SI) as a measure of the injury potential of an
impact. SI (Eqn 1) is the integral of the acceleration time curve,
weighted by the 2.5 factor observed in the Wayne State
Tolerance Curve. SI is calculated as:
dtaSI
T
∫
=
0
5.2
Eqn 1.
where a(t) is the acceleration-time pulse of the impact and T is
its duration. Equation 1 can be interpreted as “the area under
the acceleration time pulse, after the acceleration values have
been exponentiated to the power 2.5” (Fig 1B). An SI score of
1000 approximates the limit of human tolerance. Impacts with
a higher score have a greater than zero probability of causing
a life-threatening brain trauma.
HIC: The Head Injury Criterion
The purpose of the Gadd’s Severity Index SI was to express
the shock of an impact in a way that quantifies the risk of head
injury. In practice, SI scores are reasonable predictors of the
injury potential of impacts that produce focal brain injuries. For
impacts of lower intensity but longer duration (i.e those more
likely to produce diffuse brain injury), the SI calculation
produces unreasonably high values that predict more severe
injuries than those actually observed in cadaver experiments.
The Head Injury Criterion (HIC) is an alternative measure of
impact severity that is not subject to these errors. As a
measure of head injury risk, HIC (Eqn 2) is similar to SI in
principle but requires that portions of the acceleration-time
pulse be analyzed to determine the starting and ending points
that yield the highest score. The HIC score is given by:
56
−
−=
∫
=
5.2
01
01
1
0
)(
1
)(max
t
tt
t
dta
tt
ttHIC Eqn 2.
where t
0
and t
1
are the beginning and ending times of the
portion of the acceleration-time pulse being examined.
Equation 2 can be loosely interpreted as “Find the portion of
the acceleration–time pulse that has the highest average SI
score and use that as the Head Injury Criterion.”
Exponentiation of the acceleration-time pulse to the 2.5
th
power (Fig 2B) weights the accelerations according to head
injury risk using Gadd’s method; de-emphasizing lower
acceleration levels and emphasizing higher ones. The integral
(Fig 2C) accounts for the duration of the acceleration and an
iterative search finds the time interval (t
0
..t
1
) that maximises
the HIC score.
A HIC score of 1000 represents the “safe” limit of human
tolerance, above which the risk of a fatal head injury is non-
zero. The importance and validity of HIC is frequently debated
but the criterion remains extensively used. For example, in the
USA, Europe and elsewhere, government mandated
performance requirements for automotive seatbelts, airbags
and other safety devices are specified in terms of a HIC score.
It is similarly applied in the aviation industry and elsewhere. In
the sports surfacing world, HIC scores are the primary
determinant of playground surfacing shock attenuation
performance. Other specifications of surfacing shock
attenuation use a 200 g
max
limiting performance criterion, on
the basis that it approximates the HIC limit but is easier to
determine.
57
HIC scores as predictors of injury severity
Empirically determined relationships between HIC scores and
Fig 1. Example SI and HIC calculations.
(A) Acceleration-time pulse from an impact between
a surrogate head and an artificial turf surface,
showing the peak value or g
max
score.
(B) The same pulse with acceleration values
exponentiated to power 2.5. The SI score is the
area under the curve
(C) As (B) but showing the time limits, t
0
and t
1
, that
maximize the HIC score.
58
the probability of head injury (NHTSA, 1997; Prasad and
Mertz, 1985) are widely used in the automotive industry and
elsewhere as a way of estimating injury risk. Figure 2 shows
examples of “Expanded Prasad-Mertz Curves”. Each curve
estimates the probability that an impact with a given HIC score
will result in a specified level of head trauma.
Fig 2. Expanded Prasad-Mertz curves showing the relationship
between the HIC score of a head impact and the
probability of an injury
For example, consider the case of an athlete experiencing an
impact with a HIC score of 500. The curve for a “minor” injury
59
(i.e. a skull trauma without loss of consciousness) has a value
of 79% at a HIC score of 500, indicating that there is a 79%
probability that the athlete will incur a minor concussion. At the
same HIC value, the risk of a “major” injury (skull fracture,
extended period of unconsciousness) is 13%. The risk of a 500
HIC producing a critical or fatal head injury is very low, but the
probability of experiencing this head impact and not being
injured at all is only 21%.
Surface shock attenuation tests
The ultimate purpose of testing the shock attenuation
properties of a sports surface is to estimate the probability that
an impact on the surface will cause an injury. In many cases,
an absolute measure of risk is not possible and relative
measures, i.e. comparisons of the performance of different
surfaces, are commonly used.
In principle an impact test is uncomplicated. A “missile” (e.g. a
metal sphere) is dropped onto the surface, the impact is
recorded with an accelerometer embedded in the missile and
the recorded acceleration signal is evaluated. The evaluation
might include the calculation of g
max,
SI and HIC scores, for
example.
Example Impact Tests
Worldwide, there are several methods that are commonly used
to test the shock attenuation of sports surfaces.
The “Clegg Hammer” is a 2.25 kg cylindrical missile with a 5
cm face diameter that is dropped from a height of 0.46m. The
test was originally developed for testing the compaction of
road surface, but is specified in ASTM Standard F1702 as a
test of the shock attenuation of natural turf. The test method is
used for relative assessments of shock attenuation properties
and is not used to specify performance requirements.
ASTM F1936 specifies a different cylindrical missile for shock
attenuation tests of North American football fields. This missile,
60
the “F355-A” device, has a face diameter of 12.8 cm, a mass
of 9.1 kg and is dropped from a height of 0.61 m. Other tests
employ missiles with shapes that more closely resemble that
of the head. Tests of playground surfaces (ASTM F1292,
EN1177) use either a rigid headform or a hemispherical
missile dropped from various heights. Table 1 compares some
of the important properties of these test methods.
Table1: Comparison of surfacing impact test methods
Test Methods
F1702 F355-A
F1936
F1292 F1292
EN1177
Missile
Shape
Cylinder Cylinder Head-
form
Hemi-
sphere
Mass (kg) 2.25 9.10 5.00 4.60
Diameter (cm) 5.0 12.8 ~16.0 17.6
Impact Test
Drop Height (m) 0.46 0.61 Variable
Velocity (ms
-1
) 3.0 3.5 Variable
Energy (J) 10.2 55.0 Variable
Example Scores (same sample of artificial turf)
Energy (J) 10.2 55.0 54.0
g
max
80 118 251
SI 423 1630
HIC 354 1364
The dynamics of a surface undergoing an impact test are
strongly affected by the mass, shape and material properties
of the missile and by the velocity with which it strikes the
surface. Table 1 includes examples of scores from tests
performed on the same sample of artificial turf. The measured
g-max scores range from 80 g to 251 g. The HIC score from
61
an F1292 test was almost 4 times that recorded during an
F355-A test performed at approximately the same impact
energy.
Impact tests and impact simulation
Each of the impact tests described has some value as a
relative measure of shock attenuation performance. But in
order to have value as an estimator of an injury risk, the impact
test must simulate the events that present that risk. To be a
good simulator the test should mimic the structure (mass,
shape, stiffness) and dynamics (impact velocity and impact
energy) of those events.
The ASTM F1292 test method is intended to simulate the
impact between a child’s head and the surface. The
hemispherical missile or headform used in this test
approximates the mass and gross geometry of a child’s head.
The missile is dropped from a height equivalent to that of a
playground structure so the impact velocity of the test also has
good face validity.
The F355-A test method is used to test both natural and
artificial turf football fields. The impact energy (54 Joules) and
other parameters of the test are based on in-vivo head
acceleration data from a study of middle linebackers (Reid et
al, 1971). More recently, McIntosh et al (2000) found that
concussion-inducing impacts experienced by Australian Rules
Football players had a mean impact velocity of approximately
4 m s
-1
and an impact energy of 56 Joules; values that are
very close to those generated by an F355-A test (Table 1).
The geometry and inertial properties of the F355-A missile do
not represent those of the human head, however. The
differences in missile shape, curvature, mass and impact
velocity between the two methodologies have known effects
on test outcomes. For simple surface properties, these effects
can be predicted using the theory of contacting surfaces
(Johnson, 1985) and nonlinear impact models (Shorten and
Himmelsbach, 2002). The flat, circular face of the F355-A
missile compresses the surface beneath it in a uniform and
62
linear manner. In contrast, the hemispherical missile or
headform focuses the initial impact loads on a small area of
the surface. The contact area increases as the missile
penetrates the surfaces, introducing a non-linearity into the
dynamics of the impact.
In addition to differences in raw test scores, it can be shown
that the cylindrical missile introduces a bias in test results on
thin, soft surfaces. A curved head, headform, helmet, or
hemispherical missile tends to penetrate such surfaces,
bottoming them out before they can effectively absorb the
impact. The cylindrical missile engages more surface area and
applies a more uniform pressure, allowing the impact energy to
be absorbed before the thin, soft surface bottoms out.
While the hemispherical or head-shaped missile would appear
to be a better simulator of a head impact, it is still limited as a
predictor of head injury risk. The real human head has some
flexibility, which can help it absorb some impact energy. A rigid
headform does not have the same energy absorbing capacity
and, as a result, produces higher g
max
and HIC scores than a
real head.
Estimates of Head Injury Risk
In order to estimate head injury risk from impact test data, the
test scores must be adjusted to compensate for differences
between the dynamics of the impact test devices and the
human head. As an example we can consider the results of
impact tests on three different kinds of turf surface. Figure 3A
shows the typical range of g
max
scores from F355-A test of
well-maintained
natural turf, newly installed conventional
synthetic turf (carpet over a foam pad) and newly installed
,
infilled synthetic turf. (Surface conditions are emphasized
because maintenance and aging can cause test results to vary
markedly.) All would be considered to be performing in a “safe”
range because the g
max
scores are well below 200g (which
closely approximates a HIC score of 1000 HIC on this test).
63
Fig. 3: (A) Typical range of g
max
scores from 54 Joule, F355-A
impact test of three types of turf surface.
(B) HIC scores from on the same surfaces at the same
impact energy but using a rigid F1292 headform, and
adjusted to cadaver-equivalent scores.
However, once the scores are adjusted to the HIC score from
an equivalent test using a helmet-less cadaver head or
biofidelic headform with the same impact energy, a different
picture emerges. With the F355-A test’s bias in favor of thin,
soft surfaces removed, the conventional synthetic turf surface
generates higher HIC scores than natural turf and infilled
synthetic turf. Although the typical “adjusted” HIC scores
remain in the non-fatal range for a 54 Joule impact, the
conventional turf would appear to carry a higher risk of head
64
injury. This observation may help to explain why Guskeiwicz et
al (2000) observed a five times greater risk of Grade II MTBI
among football players exposed to artificial turf surfaces.
Surface design considerations
Shock attenuation
The principles underlying the influence of surface material
properties on shock attenuation performance have been
described by Shorten and Himmelsbach (2002). From the
perspective of shock attenuation, the important properties of
surfacing materials are thickness and stiffness or
compressibility. In combination, these properties determine the
energy absorption capacity of the surface and whether it can
absorb the energy of an impact without bottoming out.
Thinner surfaces must be stiffer (less compressible) in order to
absorb the same amount of energy as thicker, softer surfaces,
but are more likely to produce higher impact accelerations. For
any given impact energy, there is a minimum surface thickness
that can accommodate the impact without bottoming out; a
minimum that is independent of surface material properties.
The non-linearity of a surface’s stiffness properties is also an
important factor. If surface thickness is unlimited, surfaces that
become less compressible as the load on them increases tend
to have higher g
max
scores but lower HIC scores. Loose fill
playground surfacing materials typically reach 200g before
producing a HIC of 1000. Conversely, surfacing materials or
structures that buckle or soften when compressed tend to have
lower g
max
scores but higher HIC scores. Typically, unitary
rubber/urethane surfaces score 1000 HIC before reaching
200g.
In a more realistic realm where there is a limit on the thickness
of the surface, the best shock attenuation properties arise
when the compression of the surface is maximised during an
impact. Surfacing systems that buckle when loaded are most
65
efficient in this context. The thinnest possible surface that can
meet any given shock attenuation criterion is always one the
buckles or softens under load, rather than one that hardens as
it is compressed.
Other design considerations:
Reducing risk of head injury is not the only performance issue
that designers of sports surfaces must resolve. The frequency
of lower extremity injuries has also been linked to surface
properties. Excessive resistance to rotation between the shoe
and the surface is a known risk factor in the aetiology of knee
injuries, for example. Excessive traction may also contribute to
the occurrence of diffuse head injuries under some
circumstances (Camacho et al, 1999).
Ball bounce and roll, athlete performance, fatigue and
perception are also important design considerations. In some
instances (e.g. court sports), sports would be unplayable if the
surface was compliant enough to absorb a major impact.
Typically, athlete behaviour in these contexts is such that there
is a low risk of collision between an athlete’s head and the
surface.
Discussion
The risk of head injury is an important concern in the design of
sports surfaces. Catastrophic head injuries have life changing,
even life-threatening consequences.
Evaluating the risk of head injury in any sport is a complex
task. The context in which the injury might occur is an
important factor because the probability that the athlete’s head
will strike the surface with sufficient energy to cause an injury
varies from sport to sport. Football presents a higher risk of
head to surface contact than court sports, for example. There
is also the question of “acceptable risk” - how much risk are
the athletes, coaches, parents and the watching public
prepared to accept? Participants in sports (unlike the victims
of motor vehicle accidents, for example) choose to expose
66
themselves to potentially hazardous situations. Participants in
more aggressive contact sports would be appear to be more
risk tolerant in this regard. Finally, an individual’s susceptibility
to injury will also vary, perhaps most significantly with his or
her personal history of previous head trauma.
The shock attenuation of a sports surface is therefore only one
factor in the overall development of head injury risk. The
problem of determining this risk component directly from the
results of standard impact test remains largely unresolved.
Historically, g
max
scores of 200 g and HIC scores of 1000 have
been considered the acceptable limit on surface shock
attenuation performance. The link between the test score limits
and the cadaver impact data on which they are based in
tortuous. However, our preliminary research in this area
suggests that the conventional limits offer an appropriate level
of safety, providing the surface is capable of absorbing the
impact of a head without bottoming out. Infilled-turf surfaces,
most playground surfacing and gymnastic crash mats, for
example, meet this requirement. Conventional artificial turf is
one example of a class of surfaces that typically cannot absorb
the impact of a head without producing high g
max
and HIC
scores.
While severe head injuries to athletes are, fortunately, rare
occurrences, recent research suggests that apparently “mild”
head traumas, and especially a series of such minor
concussions can have long term, negative effects on cognitive
function. As current studies of head injury in sports are
expanded, it is probable that the head impact-specific shock
attenuation properties of sports surfaces will assume greater
importance and become a focal point of further research.
Referenced Standards
ASTM F355 Standard Test Method for Shock-Absorbing
Properties of Playing Surface Systems and Materials.
ASTM International, West Conshohocken PA, USA.
ASTM F1292 Standard Specification for Impact Attenuation of
Surface Systems Under and Around Playground
67
Equipment. ASTM International, West Conshohocken PA,
USA.
ASTM F1936 Standard Specification for Shock-Absorbing
Properties of North American Football Field Playing
Systems as Measured in the Field. ASTM International,
West Conshohocken PA, USA.
ASTM F1702 Standard Test Method for Measuring Shock-
Attenuation Characteristics of Natural Playing Surface
Systems Using Lightweight Portable Apparatus. ASTM
International, West Conshohocken PA, USA.
EN1177 Impact absorbing playground surfacing - safety
requirements and test methods. European Committee for
Standardization, Brussels, Belgium.
References
Bailes JE, Cantu RC, 2001. Head Injury in athletes.
Neurosurgery 48:26-46.
Bishop PJ, Norman RW, Pierrynowski M, Kozey J, 1979. The
ice hockey helmet: How effective is it? Phys Sports Med
7:97-106.
Camacho DL, Nightingale RW Myers BS, 1999. Surface
friction in near-vertex head and neck impact increases risk
of injury. J Biomech 32 :293-301.
Centers for Disease Control and Prevention, 1997. Sports
related recurrent brain injuries, United States. MMWR
Morbidity and Mortality Weekly Report 46:224-227.
Clarke K, Alles W, Powell J, 1978. An epidemiological
examination of the association of selected products with
related injuries in football 1975-1977: Final Report.
Bethesda MD, US, US Consumer Product Safety
Commission.
Gadd CW, 1966. Use of a weighted impulse criterion for
estimating injury hazard. Proc 10
th
Stapp Car Crash
Conference; SAE Paper 660793, Society of Automotive
Engineers, Warrendale PA, USA.
Gurdjian ES, Webster JE, 1945. Linear acceleration causing
shear in the brainstem in trauma of the central nervous
system. Mental Adv Dis 24:28.
Gurdjian ES, Webster JE, Lissner HR, 1955. Observations on
68
the mechanism of brain concussion, contusion and
laceration. Surg Gynec Obstet 101:680-690.
Guskiewicz KM, Weaver NL, Padua DA, Garrett WE, 2000.
Epidemiology of concussion in collegiate and high school
football players. Am J Sports Med 28:643-650
Johnson KL, 1985. Contact Mechanics. Cambridge University
Press, UK.
Katayama Y, Becker DP, Tamura T, Hovda DA, 1990. Massive
increases in extracellular potassium and the indiscriminate
release of glutamate following concussive brain injury. J
Neurosurg 73: 889-900.
Laforest S, Robitaille Y, Dorval D, Lesage D, Pless B, 2000.
Severity of fall injuries on sand or grass in playgrounds. J
Epidemiol Community Health, 54:475-477.
Lissner HR, Lebow, M, Evans FG, 1960. Experimental studies
on the relation between acceleration and intracranial
changes in man. Surg Gynecol Obstet 11: 329-338.
Mack MG, Sacks JJ, Thompson D, 2000. Testing the impact
attenuation of loose-fill playground surfaces. Injury
Prevention, 6:141-144.
National Highway Traffic Safety Administration (NHTSA),
Department of Transportation, 1997.FMVSS201, Head
Impact Protection, 49 CFR §571.201.
Naunheim R, McGurren M, Standeven J, Fucetola R
Lauryssen C, Deibert E, 2002. Does the use of artificial
turf contribute to head injuries? J Trauma 53: 691-694.
Patrick LM, Lissner HR, Gurdjian ES, 1963. Survival by design
– head protection. Proc 7
th
Stapp Car Crash Conference
36: 483-499.
Prasad P, Mertz HJ, 1985. The position of the United States
delegation to the ISO working group on the use of HIC in
the automotive environment. SAE Paper# 851246 Society
of Automotive Engineers, Warrendale PA, USA.
Reid SE, Tarkington JA, Epstein HM, and O'Dea TJ, 1971.
Brain tolerance to impact in football. Surg Gynecol Obstet
133: 929-936.
Saczalski, K. States JD, Wager, IJ, and Richardson, EQ, 1976.
A critical asssessment of the use of non-human
responding surrogates for safety system evaluation. SAE
69
paper # 760805. Society of Automotive Engineers,
Warrendale PA, USA 20. 150-187.
Shorten MR, Himmelsbach JA, 2002. Shock Attenuation of
Sports Surfaces. pp 152-159 in "The Engineering of Sport
IV (Ed. S. Ujihashi and S.J. Haake), Blackwell Science,
Oxford.
Schneider K, Zernicke RF, 1988. Computer simulation of head
impact: Estimation of head injury risk during soccer
heading. Int J Sport Biomech 4:358-371.
Tinsworth DK McDonald JE, 2001 Special Study: Injuries and
deaths associated with children’s playground equipment.
United States Consumer Product Safety Commission,
Washington DC, USA.
Woltys EM, Hovda D, Landry G, Boland A, Lovell M, McCrea
M, Minkoff J, 1999. Concussion in Sports. Am J Sports
Med 27:676-687.
Zhang L, Yang KL, King AI, 2001. Biomechanics of
neurotrauma. Neurological Research 23:144-156.
