THE EFFECTS OF CEREBRAL PALSY ON DENTAL DEVELOPMENT
By
Karlyn Elizabeth Lawrence

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Forensic Science – Master of Science
2013



ABSTRACT
THE EFFECTS OF CEREBRAL PALSY ON DENTAL DEVELOPMENT
By
Karlyn Elizabeth Lawrence
The ways in which cerebral palsy affects the growth and development of the
human body and skeleton have been thoroughly documented in the scientific literature.
Information on the effects of cerebral palsy on dental development, however, is scarce.
The primary goal of this thesis is to determine the effects of cerebral palsy on dental
development using a commonly practiced anthropological method of age estimation. The
Moorrees, Fanning, and Hunt (1963a,b) method of age estimation is used in this study to
estimate the ages of a sample of 51 children with a diagnosis of cerebral palsy from the
University of Michigan Health System Hospital Dentistry Clinic. The sample is
comprised of panoramic dental radiographs and periapical radiographs from 37 males and
14 females between the ages of 3 and 15.
After the estimated ages were determined, they were compared with the known
chronological ages using paired and independent t-tests to determine if a significant
difference existed between the two. The results showed that the Moorrees et al. method
of age estimation under-aged 75% of the sample of children with cerebral palsy by an
average of 1.25 years. It was concluded that the dental development of the sample of
children with cerebral palsy was delayed relative to the unaffected sample used in the
Moorrees et al. study. As a result, it was also concluded that the Moorreess et al. method
of age estimation is not recommended for use on individuals with cerebral palsy.



ACKNOWLEDGEMENTS

I would first like to thank my thesis committee: Dr. Sauer, Dr. Fenton and Dr. DeJong,
for all of their support throughout the research and writing process. Secondly, I would
like to thank Dr. Maria Regina Estrella, director of the pediatric dentistry service at the
Pediatric Dentistry Service at the University of Michigan Health System Hospital
Dentistry Clinic, who made gathering the sample for this thesis possible. Along with Dr.
Estrella, I would also like to thank her team for their hard work putting the sample
together and answering my endless questions. Lastly, I would like to thank my amazing
family: Tom, Jan, Randy, Lynn, Noel, Connor, and Emilia, for their help and patience
with me through this process. You all helped make this possible and I will forever be
grateful.




TABLE OF CONTENTS
LIST OF TABLES…………………...……………………………………………………v
LIST OF FIGURES……………………………………………………………………...vii
INTRODUCTION………………………………………………………………………...1
BIOLOGICAL PROFILE………………………………………………………....1
ESTIMATING AGE………………………………………………………………3
SUBADULT AGING……………………………………………………………..5
DENTAL AGING………………………………………………………………....5
MOORREES, FANNING, AND HUNT METHOD OF AGE ESTIMATION…..8
CEREBRAL PALSY…………………………………………………………………….11
THE EFFECTS OF CEREBRAL PALSY ON THE DENTITION……………..15
METHODS………………………………………………………………………………21
SAMPLE AND COLLECTION…………………………………………………21
ANALYSIS………………………………………………………………...…….23
STATISTICAL ANALYSIS…………………………………………………….26
RESULTS………………………………………………………………………………..28
DISCUSSION……………………………………………………………………………33
CONCLUSION…………………………………………………………………….…….42
APPENDIX…………………………………………………………………………....…46
BIBLIOGRAPHY…......................................................................................................…58




LIST OF TABLES
Table 1 – Sample by Sex……………………………………………………..……….…22
Table 2 – Sample by Formation and Resorption Methods…………………..….……….23
Table 3 – Estimation Error by Age Group………………………………...………….….30
Table 4 – Pearson’s r Correlation Between Chronological Age (CA) and Estimated Age
(EA)………………………………………………………………………………………51
Table 5 – Paired Sample t-Test Statistics for Chronological Age (CA) and Estimated Age
(EA)………………………………………………………………………………………52
Table 6 – Paired Sample t-Test Results for Chronological Age (CA) and Estimated Age
(EA)………………………………………………………………………………………52
Table 7 – Paired Sample t-Test Statistics for Chronological Age (CA) and Formation Age
(FA)………………………………………………………………………………………53
Table 8 – Paired Sample t-Test Results for Chronological Age (CA) and Formation Age
(FA)………………………………………………………………………………………54
Table 9 – Paired Sample t-Test Statistics for Chronological Age (CA) and Resorption
Age
(RA)……………………………………………………………………………………...54
Table 10 – Paired Sample t-Test results for Chronological Age (CA) and Resorption Age
(RA)……………………………………………………………………………………...54
Table 11 – Group Statistics for Independent Sample t-Test of the difference between the
Chronological Age (CA) and the Estimated Age (EA) when grouped by Formation
(Group 1) and Resorption (Group 2) Techniques…………………………………..……55
Table 12 – Results of the Independent Sample t-Test of the difference between the
Chronological Age (Ca) and the Estimated Age (EA) when grouped by Formation (Group
1) and Resorption (Group 2) Techniques………………………………………………...55
Table 13 – Paired Sample t-Test Statistics for the Chronological Age (CA) and Estimated
Age of Males (EAM)………………………………………………………………….…55
Table 14 – Paired sample t-Test results for Chronological Age (CA) and Estimated Age
of Males (EAM)………………………………………………………………………….56
Table 15 – Paired Sample t-Test Statistics for the Chronological Age (CA) and Estimated
Age of Females (EAF)………………………………………………………………...…56




Table 16 – Paired Sample t-Test results for Chronological Age (CA) and Estimated Age
of Females (EAF)………………………….…………………………...………………...56
Table 17 – Group Statistics for Independent Sample t-Test of the difference between the
Chronological Age (CA) and the Estimated Age (EA) when grouped by Males (Group 1)
and Females (Group 2)………………………………………………….………………..57
Table 18 – Results of the Independent Sample t-Test between Chronological Age (CA)
and Estimated Age (EA) when grouped by Sex……………………….………………...57




LIST OF FIGURES
Figure 1 – Stages of Deciduous Tooth Formation……………………………………….47
Figure 2 – Tooth Formation Stages and Symbols…………………………..……………48
Figure 3 – Example Graph of Moorrees et al. Norms Used to Estimate Age……………49
Figure 4 – Sample Radiograph with Moorrees et al. Stages……………………………..50
Figure 5 – Correlation Between Chronological Age and Estimated Age……………......51
Figure 6 – Estimated Error: Chronological Age and Estimated Age minus Chronological
Age……………………………………………………………………………………….53




INTRODUCTION
Forensic anthropology is the use of the principles and methods of physical
anthropology in a medico legal setting. Forensic anthropologists bring a unique
perspective to the identification of undocumented skeletal remains when the traditional
methods used by law enforcement and medicine can go no further. The forensic
anthropologist utilizes various techniques developed in a wide range of fields including
physical anthropology, archaeology, and medicine to provide physical or circumstantial
evidence pertaining to the identity of human skeletal remains and trauma in modern
criminal cases. Forensic anthropologists obtain this information by asking questions such
as: How old was this person when they died? Were they female or male? How tall were
they and of what ancestry were they? Are there any anomalies present on the skeletal
remains that cannot be explained by normal human skeletal variation? The answers to
these questions make up what is known as the biological profile and can often be a key
component in courtroom testimony. This thesis seeks to look more closely at the issue of
age estimation and more specifically, how cerebral palsy affects the rate of dental
development and the traditional methods of subadult dental age estimation.

BIOLOGICAL PROFILE
A critical part of any investigation performed by a forensic anthropologist is the
development of a biological profile. A biological profile is usually generated in the event
that a positive identification of human skeletal remains cannot be made due to a lack of
comparative information or the poor conditions of the remains themselves. Typically
included in the biological profile is an estimation of sex, age, ancestry, and stature.





Forensic anthropologist’s reports also include information on evidence for trauma,
pathologies, and skeletal anomalies (Stewart and Kerley, 1979; Iscan and Kennedy, 1989;
Buikstra and Ubelaker, 1994; Bass, 2005).
A successful positive identification often hinges on the accuracy of the
components of the biological profile. The techniques that have been developed to aid
forensic anthropologists in positive identifications must encompass a large range of
variability that exists in the human skeleton. In other words, the broader the range used
in a given estimation technique, the more likely the individual in question will fall into
that range. At the same time, however, appropriate ranges of variation must be
established in order to make each individual technique useful enough to narrow down a
possible identification and provide the most accurate estimation possible. As a result, the
techniques that forensic anthropologists use to determine age in the biological profile are
usually accompanied by prediction or confidence intervals. These confidence intervals
allow the forensic anthropologists to say statistically how confident they are that the
individual in question is encompassed in a certain range of variation.
In recent years there has also been a great push to find the most accurate and
reliable means to generate this biological profile, due to the installation of the Daubert
decision (Daubert, 1993). Before the U.S. Supreme Court issued the Daubert decision in
1993, there were no standards for determining the admissibility of evidence or expert
testimony in U.S. courts. The methods in place at the time were based solely on
relevance and general acceptance in the scientific community. The Daubert decision was
the result of the growing controversy and confusion over what should be considered
expert evidence (Dixon, 2001). It made the trial court judges the “gatekeepers” of





scientific evidence, giving them the ability to screen evidence for its relevance and
reliability based on what is known as the Daubert factors: whether it has been empirically
tested; whether it has been subject to peer review and publication; possible error rates; the
existence and maintenance of standards and controls around the technique’s operation;
and lastly its general acceptance in the scientific community (Daubert, 1993). As a
result, forensic anthropologists have been challenged with the task of scientifically
validating the methods they utilize in the development of a biological profile in order to
make them acceptable in the court of law. One important part of this validation involves
the discovery and subsequent study of environmental and biological factors that could be
affecting the development and appearance of the human skeletal system and
consequently, the estimation of all aspects of the biological profile.
This thesis seeks to closely examine this particular issue and more specifically, to
examine how a disability such as cerebral palsy can affect the development of the
dentition and how that in turn affects the usefulness of traditional anthropological
methods of age estimation when used on individuals with cerebral palsy. To accomplish
this, the Moorrees et al. (1963 a,b) method of age estimation was tested using a sample of
individuals with cerebral palsy all under the age of 16. To fully understand the extent of
these effects, however, it is first necessary to more closely examine how forensic
anthropologists normally estimates age.

ESTIMATING AGE
One of the essential elements of the biological profile is the estimation of age at
death. The job of the forensic anthropologist in determining this part of the biological





profile is to provide an age range for the skeletal remains that is as narrow and as
accurate as possible. The list of methods available to a forensic anthropologist for
estimating age at death is extensive. These methods are based on trait characteristics of
the human skeleton that change in appearance at a relatively regular rate throughout its
growth and subsequent degeneration. These regular physiological changes are broken
down into categories limited by an appropriate age range in which each change occurs,
creating a sequence of events that is directly related to a scale based on years of age. In
other words, these age estimation techniques translate the physiological age of an
individual, which is based on the appearance of the human skeleton, into a chronological
age, which is based on the number of years and months since birth (Ubelaker, 1986; Iscan
and Loth, 1989).
The various methods of age estimation can be classified into two major
categories: subadult and adult aging techniques. The aging techniques that make up the
latter category are based on the degenerative changes that occur in specific areas of the
skeleton after the individual has reached maturity. As Iscan and Loth (1989) write,
“Every bone with an open end, such as epiphyseal regions, sutural borders, and articular
surfaces, will show visible signs of aging.” In addition, the visible signs of aging that
progress in each of these locations are unique and follow a predictable pattern. Most of
the studies on the adult aging process, and the resulting age estimation techniques, are
focused on four major areas of the skeleton: the pubic symphysis, the sternal rib ends,
cranial sutures, and the auricular surface (Krogman, 1962; Stewart, 1979; Iscan and Loth,
1986; Bass, 2005).





SUBADULT AGING
The second category is subadult aging, which is based on the formative changes
of the human skeleton that occur during the growth and development of the subadult
skeleton. The major formative changes that are utilized most often in subadult aging
include the appearance of ossification centers, the length of long bones, epiphyseal union,
and the development of the dentition (Ubelaker, 1986). The first three aging techniques,
although very useful, are all dependent on skeletal development. The growth and
development of the skeleton has been shown to be susceptible and exhibit a great degree
of variation due to environmental influences, populational differences, health
deficiencies, and nutritional factors. The techniques based on the development of the
dentition, however, have been shown to provide more accuracy and reliability due to the
fact that dental tissue seems to be buffered to a larger degree against the variation caused
by the same stressors (Demirjian et al., 1985; Demirjian, 1986; Smith, 1991; Cardosa,
2007). As a result, the most preferred methods of subadult age estimation among
anthropologists are those that utilize the developing dentition (Ubelaker, 1986).

DENTAL AGING
Aging techniques based on the development of the dentition can be broken down
into two categories: dental eruption and tooth formation. Dental eruption has been used
since the mid 1800’s as a measure of child maturity and since that time, hundreds of
studies have focused on the relationship between dental eruption and age (Liversidge et
al., 1998). In these early years, dental eruption was a commonly used measure of
biological age mainly due to its ease of use. Researchers and practitioners had only to





look inside the child’s mouth to observe the state of dental eruption versus taking any xrays or measurements. Schour and Massler (1941) studied the pattern and rate of dental
eruption and development in a large sample of individuals encompassing children as
young as five months in utero up to adults 35 years of age. The authors developed a
timeline that was delineated by specific age categories and defined by the expected
amount of dental calcification and eruption during that period of life. Although Schour
and Massler (1941) studied both calcificaton and eruption, the focus was on eruption
which in itself is difficult for anthropologists to use due to the fact that that the eruption
of the teeth is based on their emergence through the gum, which is usually no longer
present on skeletal remains (Ubelaker, 1986).
Another good example of a study on dental eruption rates is that of Gron (1962).
Gron (1962) studied and compared the rates of dental eruption, dental development,
skeletal development, and chronological age in order to try to narrow the time interval
estimates of dental eruption for the purpose of timing orthodontic treatment in children.
Like most of the other studies of dental eruption of the time, Gron found that tooth
emergence through the gingival is more closely associated with root formation then it is
with chronological age. Although Gron and other similar studies provide information on
the average eruption times of their samples, the overall eruption rates seem to conflict
with each other and encompass a great amount of variation (Ubelaker, 1986).
Tooth formation, on the other hand, is a popularly used growth marker in
estimating age due to the fact that it has the highest time stability of all of the growth
systems in the body (Liversidge et al., 1998). Among forensic anthropologists it is
regarded as one of the most accurate and reliable methods of calibrating growth and





consequently, determining age at death (Ubelaker, 1986; Smith, 1991). Like estimates of
age based on weight, height, and skeletal development, dental age is a measure judged in
terms of physiological development. The human dentition has a specific and predictable
pattern of growth that begins in utero with the formation of the cusps of the deciduous
dentition and comes to completion with the closing of the root apices as the skeleton
nears maturation (Smith, 1991).
This pattern of growth has been studied extensively dating back to the late 19th
century with the first published descriptions and illustrations detailing the timing and
appearance of tooth formation. These early works were based mostly on dissections and
the visual inspections of specimens and as a result, were grossly simplified (Smith, 1991).
Later works on dental development were generally lumped together with dental eruption
and based loosely on the work of Schour and Massler (1941), which used growth charts
and descriptions of the entire dental growth process. These growth charts were extremely
popular and laid the foundation for numerous future studies including a landmark study
by Lewis and Garn (1960) where the researchers found that the stages of tooth formation
had lower coefficients of variation than the stages of skeletal development using a sample
of healthy middle class children. The results of this study, and many others since its
publication, have indicated that dental development provides the best and most accurate
means of determining age at death due to the predictability and reliability of the stages of
formation and resorption (Ubelaker, 1986, Smith, 1991).





MOORREES, FANNING, AND HUNT METHOD OF AGE ESTIMATION
One of the most popular methods of sub-adult age estimation, and consequently
the one used in this thesis, comes from a two-part study by Moorrees, Fanning, and Hunt
(1963 a,b). In the first of the two studies (Moorrees et al., 1963a), the process of the
formation and resorption of three deciduous teeth is detailed. The method itself is based
on attributing arbitrary stages to the continuous process of dental growth. The purpose of
their publication was not only to clarify this hard to observe process, but also to present a
new method for determining physiological age in children. Moorrees et al. (1963a)
observed the pattern of formation and resorption of the deciduous canines and the first
and second deciduous molars through the use of oblique and lateral radiographs of 136
boys and 110 girls. A series of arbitrarily defined stages were created for both tooth
formation and root resorption. The stages of tooth formation begin with the “initial
formation of the crown” and end with the “closure of the apices of the roots.” Moorrees
et al. identified a total of 12 stages of formation for the deciduous mandibular canines and
13 stages for the deciduous molars. In addition, three stages were identified for the
resorption of the deciduous canines and molars and are described by Moorrees et al. as
being chosen, “for studying the exfoliation process from its earliest beginning, consisting
of formation stages in reverse.” Moorrees et al. represent each stage of formation and
resorption with a hand drawn picture that best represents the appearance of the tooth at its
particular stage of development (Appendix Figure 1). A written description is also
included in the publications (Appendix Figure 2).
The results of the study by Moorrees et al. (1963a) were presented graphically in
their publication and separated into male and female categories due to the differences





between the development of the different sexes. The authors created a series of charts
organized by sex and tooth type (Appendix Figure 3). Within each chart, a given
developmental stage and its corresponding tooth are represented by a horizontal bar that
is placed under a time-line delineated by years of age. When a particular stage of
development is identified on an individual tooth, the average age in years of the child and
the appropriate standard deviation can be determined by finding the appropriate
horizontal bar and referencing the time line. Moorrees et al. encourage averaging the
results of several teeth to attain a more accurate estimation of age.
The second study by Moorrees et al. (1963b) extended the research of the first
publication by including the formation stages of ten permanent teeth. The sample used in
this study included the dental radiographs of 48 male and 51 female children from the
Stuart material in Boston and an additional 136 white male children and 110 white female
children, which were gathered by Dr. Arthur B. Lewis of the Fels Research Institute in
Yellow Springs, Ohio. The teeth utilized in this study included the incisors of both the
mandible and maxilla as well as the mandibular canines, the first and second premolars,
and the first, second, and third molars. As was done for the original study, a series of
arbitrarily defined developmental stages were created for the formation of the previously
mentioned ten teeth. Each stage is represented both by a picture description for easy
comparison and graphically on a corresponding chart referencing the corresponding age
estimate just as it was done in the original study. The authors defined 13 stages of
development for the single rooted teeth and 14 stages for multiple rooted teeth, both
beginning with the initial appearance of the crown and culminating with the closing of
the root apices. Age is estimated in the same way it was in the original study, by





referencing the correct chart, locating the estimated stage of development, and then
referencing the time line.
The method of subadult age estimation developed by Moorrees et al. (1963 a,b)
has become one of the most widely used and greatly preferred techniques by forensic
anthropologists in North America for a variety of reasons. First of all, the samples
utilized in both studies are evenly distributed over each age group, ensuring that each age
group is well represented (Smith, 1991). Secondly, the Moorrees et al. studies are one of
the few that can accurately be used for age estimates extending back to the “birth” time
period. This widens the usefulness of the technique to include age estimates from birth
all the way up until the early 20s (Smith 1991).
Moorrees et al. caution, however, that the age estimates created by this technique
can be influenced by several factors. One of the factors includes the quality of the
radiographs used to determine the stage of development. Radiographs that are difficult to
read may mask the true stage of development of a particular tooth. In addition, the
experience of the observer can also have a negative impact on the accuracy if the
observer has had very little experience with this type of age estimation.
Since the publication of their aging method in the 1960’s, a few other issues have
surfaced pertaining to the use of this technique. The major issue surrounds the
composition of the sample used in the study, which was made up of exclusively healthy
white children. As a result, the developmental norms that were established pertained to
this specific population and children that were from other populations did not fit as neatly
into their well-defined stages (Moorrees et al., 1963b; Scheuer et al., 2000; Scheuer et al.,
2004). Harris et al. (1990), for example, found that the Moorrees et al. method under-





aged a sample of male and female black children from the middle southern United States.
Interestingly, this study found that mineralization occurs significantly earlier (almost 5%
earlier) in the black population versus the white population that was used in the study.
More recently, Phillips et al. (2009) tested the Moorrees et al. method on a large and
ancestrally diverse sample of children, including both black and white South African
children and Indian children. The results of the study showed that the Moorrees et al.
method of age estimation consistently under-estimated the age of the children in the
sample. The authors concluded that the Moorrees et al. method was not applicable to
South African populations and new standards should be created specifically for them.
These two studies bring to light the ever-important issue of validation and
understanding both the strengths and weaknesses of the methods we use in forensics. As
stated earlier, due to the nature of the forensic sciences and what they are used for, it is
paramount that the limitations of our methods are fully understood. Differences caused
by sex, population, and the environment can all play a role in the accuracy and
dependability of any technique used in the forensic sciences. It is the goal of this thesis
to look at one small part of this complicated picture by testing the Moorrees et al. method
on a sample of individuals with cerebral palsy. This goal goes a step further with the
hopes of ascertaining a better understanding of the development of the dentition of
children with cerebral palsy. To do this, however, it is necessary to take a closer look at
cerebral palsy itself and its affect on the body and the dentition.

CEREBRAL PALSY
Cerebral palsy (CP) is a disease that is known to cause a delay in the growth and
development of both the skeleton and the dentition (Megyesi et al., 2009) and is also the




most common of the central nervous system disorders found in children (Moller 1973).
CP is a collection of non-progressive neurologic disorders caused by permanent damage
to the brain during the time when the central nervous system is maturing. Because the
damage occurs specifically to the motor control centers of the brain, CP most notably
affects muscle coordination and voluntary muscle control and as a result, the most
common symptoms of this disorder include muscle stiffness or paralysis, involuntary
movements, and poor balance (Miller, 2005; Pellegrino, 2002).
The etiology of a specific case of cerebral palsy can be difficult to elucidate due
to the wide range of conditions that cause CP. Basically, CP results when something
interferes with cerebral development and physiology before or shortly after birth. These
conditions include and are in no way limited to congenital developmental deformities,
prematurity, birthing problems, hypoxia, intrauterine infection, and postnatal conditions
such as trauma, infections, and toxicities (Miller, 2005). The occurrence of CP has
remained fairly steady at around 1.4-2.4 cases per 1,000 since the 1970’s. The
percentage of children born prematurely with cerebral palsy, however, has risen steadily
within the same time frame while incidences of CP caused by problems associated with a
traumatic birth experience have fallen dramatically (Pellegrino, 2002).
As stated earlier, cerebral palsy is a group of disorders that are generally classified
according to the type of motor impairment that results from damage to different regions
of the brain. The range of disability and severity of the symptoms caused by CP vary
greatly between individuals. In the mildest cases the effects on movement and muscle
control can be almost imperceptible in an individual. On the other end of the spectrum
individuals with severe forms of CP can lose all control over muscle movement





throughout the body, a condition that leaves them completely disabled (Pellegrino, 2002;
Weddell et al., 2004). In the most severe cases, the entire musculoskeletal system can be
affected resulting in what is known as total body cerebral palsy.
The group of disorders that make up the umbrella term cerebral palsy can be
classified into four types. The most common type of CP is the spastic type, which makes
up nearly 70% of all cases (Weddell et al., 2004). Spastic CP occurs when the damage to
the brain is focused in the area of the cerebral cortex (Moller 1973). Spastic CP can be
further broken down into three additional categories based on the location and extent of
the damage in the brain. Spastic hemiplegia, for example, occurs when only one side of
the brain is damaged resulting in the dysfunction of the opposite side of the body,
generally both arm and leg. Spastic diplegia, on the other hand, involves damage to both
sides of the brain and results in loss of function in both legs and a minor loss of function
in the arms. The most severe form of the spastic type is spastic quadripelgia. Loss of
muscle control is almost universal throughout the body, involving the four limbs, the
trunk, and possibly the mouth, tongue and pharynx. This last type of spastic CP is
generally accompanied by a wider spread cerebral dysfunction that can include symptoms
such as seizures, mental retardation, and other symptoms pertaining to the mouth and
dentition that will be discussed later (Pellegrino, 2002). The most common symptoms
associated with the more severe expressions of spastic cerebral palsy include exaggerated
contraction of muscles, limited control of the neck muscles, and the inability to control
the oral and masticatory musculature (Weddell et al., 2004; Troutman et al. 1982).
The second type of CP, comprising approximately 15% of all cases, is known as
dyskinetic and is characterized by changing patterns of muscle tone throughout a given





day (Weddell et al., 2004). Generally, this type of CP affects the entire body and
symptoms can vary between involuntary rigid or jerky movements to smoother more
constant writhing movements. Individuals who are affected by this type of cerebral palsy
also exhibit a lack of control over the facial muscles causing excessive drooling, severe
bruxism and difficulties with chewing and swallowing (Pellegrino, 2002).
The third and most infrequent type of CP is ataxic cerebral palsy and only occurs
in approximately 5% of all cases. Like dyskinetic CP, ataxic generally involves the entire
body. It is characterized, however, more by abnormalities involving balance and the
positioning of the trunk of the body and its limbs. As a result, individuals with ataxic CP
generally exhibit an abnormal gait and tend to stumble frequently. Difficulties are also
experienced with the motor control of the hand and arm when trying to grasp objects.
Tremors in the hands and arms are also common as well as a general uncontrollable
trembling throughout the entire body (Weddell et al., 2004; Pellegrino, 2002).
The final classification, mixed cerebral palsy, is diagnosed when more than one
type of motor dysfunction is observed. It occurs in approximately 10% of all cases and
presents when multiple areas of the brain are damaged. The symptoms can include any
combination of those mentioned above depending on the extent and severity of the
damage (Weddell et al., 2004; Pellegrino, 2002).
Because the damage to the brain that causes cerebral palsy occurs during the
brain’s development and is generally severe, cerebral palsy very rarely occurs in isolation
to other disabilities. Individuals with cerebral palsy are also very likely to have other
impairments that originate from the same damage to the central nervous system (Low et
al., 1982). Mental retardation is one such co-morbidity that is quite common among





affected individuals occurring in approximately 60% of all cases. Seizures are also quite
common, occurring most frequently during the infant and early childhood years (Low et
al. 1982; Weddell et al., 2004). Both hearing and vision can also be affected in several
different ways with the incidence of a condition known as strabismus being the most
common issue. Strabismus is the inability of the extraocular muscles to work in
coordination with each other resulting in the misalignment of the eyes. Lastly, speech
problems are prevalent among individuals with CP occurring in approximately half of all
reported cases. This high percentage is attributed to the lack of control over the muscles
associated with speech (Low et al., 1982; Weddell et al., 2004).

THE EFFECTS OF CEREBRAL PALSY ON THE DENTITION
The effect that cerebral palsy has on the general status of the dentition has been
well documented (Pope and Curzon, 1991; Rodrigues et al., 2003). Like unaffected
children, children with cerebral palsy suffer from dental caries, periodontal disease,
enamel hypoplasia, bruxism, and trauma although many of these conditions may be
severely exaggerated (Franklin et al., 1996). The development of the teeth of individuals
affected by cerebral palsy, however, has largely been ignored in the literature. This may
be partly due to the fact that the dental treatment of children with CP is often difficult to
accomplish due in part to the child’s inability to control the musculature of their head and
neck. As a result, general dental care and dental radiography may be lacking in
individuals with CP.
The literature on cerebral palsy and how it affects dental development is limited,
however, research on the dental status of children with cerebral palsy and case reports





documenting the abnormal eruption of their teeth are common. In two separate studies
by Pope and Curzon (1991) and Rodrigues et al. (2003) the general dental status of a
sample of children with cerebral palsy was compared with a control sample of unaffected
children. In both studies, the children with cerebral palsy presented with higher rates of
attrition as well as a significant delay in the eruption and shedding of primary teeth.
Rodrigues et al. (2003) also observed significantly higher percentages of malocclusion
and bruxism among the affected children with permanent dentition.
Staufer et al. (2009) found similar results in their case report on the failure of
tooth eruption in two children with cerebral palsy. Both patients presented with mixed
dentition and abnormal amounts of attrition. In both cases the eruption of the permanent
dentition began at age 17 only after several of the primary teeth were extracted. Tooth
eruption proceeded at a slow rate as the corresponding primary teeth were removed.
Interestingly, Staufer et al. (2009) also observed that the mineralization of the
permanent dentition was within the normal range of timing, however, the resorption of
the primary teeth failed in most cases, especially where there had been extreme tooth
wear. The authors ascribe this issue to the hyperactivity of the odontoblasts caused by
the pathological wear of the teeth. This abnormal wear was attributed to the severe
bruxism experienced by the two children with cerebral palsy. Staufer et al. (2009) states
that these over active odontoblasts continue to lay down secondary dentin and, in some
cases, this can result in the replacement of the pulp chamber and root canal orifices by
secondary dentin. As a result, the individual tooth fails to begin the process of resorption
and exfoliation.





As stated earlier, individuals with cerebral palsy generally exhibit high rates of
attrition due partly to the inability to control the muscles of the face, which can result in
bruxism. Another possible cause of the high rates of dental erosion that are observed in
children with CP is gastrointestinal reflux. In a study by Shaw et al. (1998), researchers
looked at the relationship between dental erosion, cerebral palsy, and gastrointestinal
reflux using a sample of 21 individuals with CP and 30 unaffected individuals. What
they found was that there was a strong relationship between dental erosion and
gastrointestinal reflux and less of a relationship between dental erosion and the bruxism
caused directly by cerebral palsy. This finding is significant due to the fact that
gastrointestinal reflux is an issue that commonly occurs along with a diagnosis of
cerebral palsy and may be an important contributing factor in the high frequency of tooth
wear observed in children with cerebral palsy, which is a challenge to the more
traditional explanation of bruxism (Shaw et al. 1998).
In addition to attrition, bruxism, and delayed eruption; another commonly
encountered dental issue that often accompanies cerebral palsy is malocclusion. The
muscles of the face and jaw play a major role in the development of facial growth and
overall occlusion. Because children with cerebral palsy generally have problems with
muscle tone in these areas, open bites and over jets are common. In a classic study by
Rosenbaum et al. (1966), the authors studied the occlusion of 124 children with cerebral
palsy and compared it to the occlusion of their control sample of 141 unaffected children.
Interestingly, they observed little to no statistical increase in the occurrence of open bite,
over jet, and overbite in their sample of children with CP. Although the observed
percentages were the same or only slightly higher for the children with cerebral palsy, the





actual measurements of the over jet, overbites, and open bites were significantly greater
than the unaffected children being on average 0.8 mm greater, 0.5 mm greater, and 1.7
mm greater respectively.
In two more recent studies, both Franklin et al (1996) and Carmagnani et al.
(2007) found an increased prevalence of malocclusion in their samples of children with
cerebral palsy as compared to samples of unaffected children. Franklin et al (1996)
found that their small sample of children with CP had both an increased amount of
overbite and over jet. They postulate that the malocclusion experienced by CP patients
may be caused by both lip position and lip competence but caution that the sample used
in this study is probably too small to draw any firm conclusions. Carmagnani et al.
(2007), on the other hand, studied a larger sample of 104 individuals with a clinical
diagnosis of cerebral palsy in order to observe their state of dental occlusion. The sample
was broken down by type of cerebral palsy and the authors found that the incidence of
malocclusion differed slightly by type of CP. The spastic type, for example, showed the
highest incidence of malocclusion with a high frequency of both open bite and class II
malocclusion (overbite). The ataxic group presented with the lowest frequency of
malocclusion as compared to the other groups, however, both the ataxic and the
dyskinetic groups demonstrated a higher than normal frequency of deep overbites. The
authors concluded that the cause of the Class II malocclusion observed throughout all
types of CP can be attributed to the abnormal craniofacial growth and development that
CP patients go through.
In their review of the dental literature, Bhat and Nelson (1989) examined the
frequency of dental enamel defects in children with neurologic disabilities, including





cerebral palsy. They found that children with CP were twice as likely than normal
children to have developmental enamel defects in their primary dentition. The authors
attribute this high frequency to the sensitivity of the developing tooth germ. During its
growth, the tooth germ is susceptible to a wide range of developmental disturbances
including those that cause CP. These disturbances leave permanent defects on the teeth
and record a physical history of insults that occurred during prenatal and early postnatal
life. The authors believe that this history could possibly be used to determine the specific
timing of the insults and thus help to elucidate the complicated etiologies of cerebral
palsy. Unfortunately, the authors conclude that the existing literature has not adequately
tested this idea or has even begun to scratch the surface of properly describing the
specifics of these enamel defects.
More recently, Keinan et al. (2006) examined the microstructure of teeth from
children with cerebral palsy and Down syndrome (DS). The results of their study showed
that children with CP and DS lay down a significantly less amount of enamel prenatally
than normal children. They also found that the enamel of the mesial cusps among DS
and CP children were less highly mineralized than unaffected children. The differences
in mineral content and enamel formation demonstrate an impaired function of the
ameloblasts in children affected by DS and CP. The effects of the impaired function of
the ameloblasts in CP cases were found to be relatively short in duration and normal
mineralization returned in the subsequent stages of development. The authors concluded
that future research should continue to look at the relationship between episodes of stress
during pregnancy and enamel defects and how they can be used to reconstruct the order
and severity of systemic insults to the developing fetus.





Throughout the scientific literature, only two studies focus on the relationship
between the chronological age and the dental developmental age of individuals with
cerebral palsy. Megyesi et al. (2009) report on a case where several traditional forensic
anthropological methods of age estimation significantly underestimated the age of an
individual with a known history of cerebral palsy. Using the states of epiphyseal fusion
throughout the individual’s skeleton, age was estimated to be between the ages of 11 and
15. The true chronological age of the male individual, however, was between the ages of
21 and 23. Dental age estimates using Moorrees et al. (1963a,b) and Smith (1991) were
slightly closer to the chronological age being 16.4 years on average. Megyesi et al.
(2009) attributes this discrepancy in estimated age and chronological age to be partly
caused by malnutrition and the difficulties associated with feeding and treating patients
with severe cases of CP.
Ozerovic (1980) briefly discusses a study that tested for a correlation between
chronological, dental and skeletal age. Using a sample of children with cerebral palsy,
the author documented their skeletal age using radiographs of the wrist and hand and
their dental age using Nolla (1960). The results of the study were split into two
categories based on the type of cerebral palsy: dyskinetic and spastic cerebral palsy. The
results showed that both dental and skeletal development are affected by cerebral palsy.
However, it also concluded that the rate of dental and skeletal development was more
affected in the individuals with the dyskinetic type. Both dental and skeletal
development showed a low correlation with each other and with chronological age in this
type. In the spastic type of cerebral palsy, however, the results showed that skeletal and





dental development generally followed each other but neither could be correlated with
chronological age.
One of the common themes shared by all of the previously mentioned research is
the lack of information and research on the effects of cerebral palsy on the dentition in
general. Rosenbaum et al (1966) complain specifically of the wealth of information on
the diagnosis of cerebral palsy and yet the near absence of any practical information
pertaining to the treatment and management of the problems associated with the dentition
of individuals with cerebral palsy. In recent years there has been a surge of research
filling these gaps in information that Rosenbaum et al. complained about; however, the
research into the development of the dentition itself is still quite limited. The timing of
the dental growth and development of children with cerebral palsy is only mentioned
briefly throughout the scientific literature, usually as a secondary observation. This thesis
seeks to look specifically at this gap in the scientific literature by comparing the dental
development of a sample of children with cerebral palsy with the study by Moorrees,
Fanning, and Hunt (1963a,b). More specifically, this thesis proposes to examine the
effects of cerebral palsy on the development of the dentition and how those effects can
influence the reliability and accuracy of the traditional forensic dental methods of age
estimation.

METHODS
SAMPLE AND COLLECTION
Panoramic dental radiographs and periapical radiographs of 77 individuals who
have a diagnosis of cerebral palsy were identified at the Pediatric Dentistry Service at the





University of Michigan Health System Hospital (UMHS) Dentistry Clinic under the
supervision of Dr. Maria Regina (Ninna) Estrella, director of the pediatric dentistry
service. An approved UMHS staff member randomly numbered each radiograph,
removed all identifiable information, and recorded both the sex of the individual and the
chronological age of the individual at the time the radiograph was taken. The
radiographs were then scanned into an electronic database using the randomly assigned
number and without any identifying information so that dental age assessments would be
completely blind. All other information needed for this study was later entered into a
second electronic database for comparison.
The sample originally consisted of 77 individuals. Those with co-morbidities,
such as epilepsy and developmental delay, were included in this study due to the
difficulty of isolating patients with only a single diagnosis of cerebral palsy. After
reviewing each patient file, 26 individuals were excluded from the study due to either
poor radiograph quality or the absence of any information on the radiograph that would
lead to an age assessment using the method developed by Moorrees et al. (1963 a,b). The
remaining 51 individuals in the sample that were used for this study included 37 males
and 14 females between the ages of 3 and 15 years old. These particular files contained
at least one clear radiograph that displayed a tooth that could accurately be identified and
fell within the qualifications for age assessment laid out by Moorrees et al. (1963 a,b).
Table 1 – Sample by Sex
Male

Total

37
73%



Female
14
27%

51
100%



Table 2 – Sample by Formation and Resorption Methods
Formation

Resorption

Total

45
83%

9
17%

54
100%

The radiographs were scanned into a personal computer using a Canon scanner
and the Cano Scan 4.1 software. The software was set on a grayscale for black and white
positive film. Image quality was set at the software’s standard 400 dpi and the resulting
images reproduced by the scanner were saved as jpg files. After all of the files were
scanned, each radiograph was opened in the photography software Picasa in order to
enlarge the image and improve the image’s contrast. This allowed the observer to attain
the highest amount of picture clarity without physically altering the radiographs in any
way.

ANALYSIS
To determine the dental age of each individual, the method developed by
Moorrees et al. (1963 a,b) was used due its common use among forensic anthropologists.
The Moorrees et al. method also utilizes a wide range of teeth that can be used together to
attain a more accurate average age or can be used on isolated teeth in the event there is
only one tooth available for age estimation. Due to the difficulty associated with
radiographing the teeth of children with cerebral palsy, the latter point was very
important in this study.
To determine age using the two studies by Moorrees et al. (1963 a,b), it was first
necessary to identify what teeth were available for observation in the radiographs of the





current sample and which of those teeth were utilized in the Morrees et al. aging
technique. The first study by Moorrees et al. (1963a) focuses on the formation and
resorption of the deciduous mandibular canine, first molar, and second molar. In this
study, Moorrees et al. outlined 12 stages of formation for the canine and 13 stages for the
molars, beginning with the first appearance of the crown and ending with the closing of
the root apex. Moorrees et al. also outlined three stages of root resorption which were
used in the current study (Appendix Figure 1). The stages of resorption were far less
specific than the stages of formation and were defined by Moorrees et al. simply as root
resorbed ¼, ½, and ¾. Each stage of formation and resorption was represented by a brief
written description as well as accompanied by a picture description to aid the observer in
making an accurate comparison. Both the written descriptions and the drawings were
used in the current study to make a comparison.
In the second publication in 1963, Moorrees et al. (1963b) extended their dental
development study to include the stages of formation for ten permanent teeth that were
also used in the current study. These were the maxillary and mandibular central and
lateral incisors, the mandibular canines, the mandibular premolars, and the first, second
and third mandibular molars. In this study Moorrees et al. outlined 13 stages of
formation for the single rooted teeth and 14 stages of formation for the mandibular
molars. The stages of formation for the permanent teeth begin with what the authors
described as the “initial cusp formation” and ended with the closure of the root apex or
apices. Once again the stages were all represented by both a written description as well
as a simple picture description and both were used in the current study to aid in age
estimation.





In the event that one of the previously mentioned teeth was available for
observation in the current study, the radiograph was observed in digital form at both its
actual size and various other magnifications to attain the greatest image clarity. Once the
tooth was observed in its clearest state, it was then compared with the graphic
representations outlined by Moorrees et al. and the corresponding stage was recorded
(Appendix Figure 4).
Once the stages were determined for each individual, it was possible to begin the
estimation of age using Moorrees et al. To start this process a database was created that
included each individual’s randomly assigned number, their sex, and the corresponding
stages of dental development that were identified in this study. After the database was
complete the age of each individual was calculated using the charts of formation and
resorption created by Moorrees et al. These charts are separated by sex,
formation/resorption, and deciduous/permanent teeth. Each chart is made up of a series of
horizontal bars that each represents one of the stages of development. Each horizontal
bar then contains the average age of the individuals at the specified stage plus or minus
one and two standard deviations. The estimated age is found by determining what stage
of development the tooth is in, locating the proper chart to use, finding the correct
horizontal bar that represents the proper stage of development, and finally drawing a
perpendicular line through the horizontal bar that reaches the time lines found at the top
and bottom of each chart. The number on the time line that the perpendicular line passes
through is considered the estimated age.
For this particular study, several procedures were followed. Age was documented
in quarters of a year. For example, ages were recorded as 8.0 years, 8.25 years, 8.50,





years or 8.75 years. If the perpendicular line crossed the time line between the quarter
year marks, the age was rounded up to the next quarter year. If multiple teeth were
available for study on a single individual, age was determined separately for each tooth
and then the resulting ages were averaged together to attain the overall estimated age as
was recommended by Moorrees et al.

STATISTICAL ANALYSIS
After all available information was collected, a database was created for statistical
analysis. This database included all information collected in the current study: individual
number, sex, chronological age, formation age, resorption age, and estimated age. The
goal of the statistical analysis was to determine the correlation between chronological age
and dental age among children with cerebral palsy and to determine the reliability and
accuracy of the Moorrees et al. method of assessing age using tooth formation and root
resorption on individuals with cerebral palsy. All statistics were performed using the
SPSS statistical software.
The statistical analysis began with calculation of the correlation between
chronological age and the estimated ages using a Pearson’s r correlation. After the
correlation was found, paired t-tests were run to determine the differences between the
means of the chronological ages and the estimated ages. The paired t-test was used in
this situation because it specifically compares the means of two populations that are in
some way related to each other (Moore et al. 2003). In the case of this study, both
estimated age and chronological age are dependent on development and are therefore
related.





Next, the average estimation error was calculated and represented graphically.
The sample was then broken down into three age categories in order to gain a better
understanding of where the majority of this estimation error was occurring. The three
age categories were: 3-6 year olds, 7-10 year olds, and 11-15 year olds.
After the entire sample was statistically analyzed, it was broken into smaller
groups. The results of the following analyses must be interpreted cautiously due to their
extremely small sample sizes. The first two breakdowns of data in this study were
between estimated ages determined by formation and those that were determined by
resorption. To begin with, a paired samples t-test was performed between the
chronological age and the formation age to determine if there was a difference between
the means of both samples. The paired samples t-test was then repeated for the
chronological ages and the formation ages to attain the same information.
Lastly, an independent samples t-test was used to determine the differences in
means between the estimated age and the chronological age (estimated age minus
chronological age) and to determine if the sample was being under or over aged by the
formation and resorption techniques. The independent samples t-test was used in this
study due to its ability to measure the difference of means between two independent
groups.
The data were also broken down by sex to discover if there was any difference
between the estimated ages and the chronological ages when categorized by sex. Paired
samples t-tests were run to determine the differences in means between the chronological
age and the estimated ages of males and females independently. These tests were then
followed by an independent samples t-test to further analyze the difference between





males and females and to better understand if any systematic over or under aging was
occurring.
The purpose of this thesis is to examine the effects of cerebral palsy on the
development of the dentition and how those effects can influence the reliability and
accuracy of the Moorrees et al. method of age estimation. The specific aims are twofold:
(a) to determine the correlation between chronological age and dental age among children
with cerebral palsy and (b), to determine the reliability and accuracy of the Moorrees et
al. method of assessing age using tooth formation and root resorption on individuals with
cerebral palsy. It is hypothesized that the sample of children with cerebral palsy will
show a rate of dental development that is delayed relative to the sample of unaffected
children used by Moorrees et al. As a result, the Moorrees et al. method will be unable to
accurately estimate the ages of the children with cerebral palsy.

RESULTS
The statistical analysis began by calculating the correlation between the
chronological age and the estimated age using the method created by Moorrees et al.
(1963 a, b). The r-value calculated using this test was .947, and was found to be
significant at the .01 level (Appendix Table 4). These results indicate a strong positive
correlation between the chronological age and the estimated age. This strong positive
correlation demonstrates that as the chronological age increases, so does the estimated
age. The results of this test and their correlation are demonstrated graphically in
Appendix Figure 5.





Second, a paired t-test was run to measure the difference between the means of
the chronological age and the estimated age in this study. The significance level was set
at .01 and both the null and alternative hypotheses were determined. The null hypothesis
was that there was not a statistically significant difference between the chronological ages
and the estimated ages. If no statistically significant difference was found the null
hypothesis would be accepted and it would be concluded that the rate of the development
of the dentition of the children with cerebral palsy in this study is statistically similar to
the unaffected children in the Moorrees et al. study. However, if a statistically significant
difference was found, then the null would be rejected. The alternative hypothesis was
that the rates of development were indeed different and the Moorrees et al. method is not
appropriate for use on this particular sample due to the rates of development being
significantly different.
The means calculated by the paired t-test were 8.26 for the chronological age and
7.43 for the estimated age (Appendix Table 5). A comparison of these numbers indicates
that the Moorees et al. method under-aged the sample as a whole. The t value was
calculated as 4.97 at 50 degrees of freedom with a p value of .00 which can also be
interpreted as less than .01 (Appendix Table 6). Because the p value was calculated at
less than .01, this demonstrates that there is indeed a statistically significant difference
between the means of the chronological age and the estimated age. As a result the null
hypothesis was rejected and the alternative hypothesis was accepted and it was
determined that the rate of development of the dentition of the children with cerebral
palsy in this study was different than the sample of unaffected children used by Moorrees
et al.





To better understand this difference, the data were used to test the degree of error
between the chronological ages and the estimated ages. It was found that the Moorrees et
al. method under-estimated the ages of 75% of the sample by an average of 1.25 years.
As was stated earlier, the range of chronological ages used in this study was 3 to 15
years. The estimated ages produced using the Moorrees et al. method, however, ranged
from an under estimation of 4.5 years on the low end to an over estimation of 1.13 years
on the higher end. It was also found that 12 individuals out of the total 51 (nearly 24% of
the sample) were under-aged by two or more years. The majority (nine individuals) of
these under-aged children were older than ten years while the remaining three were
between three and nine years of age. The overall estimation error was represented
graphically in Appendix Figure 6 by comparing the chronological age to the difference
between the estimated age minus the chronological age.
To gain a clearer picture of the degree of under aging that was occurring, the
sample was broken down into three different age groups: 3-6 year olds, 7-10 year olds,
and 11-15 year olds. The following table demonstrates the average number of years each
age group was under-aged by. The information in table three clearly demonstrates that the
majority of the under-aging is occurring in the oldest age category, individuals 11 years
and older.
Table 3 – Estimation error by age group
3-6 year olds

11-15 year olds

(20 individuals)

(18 individuals)

(13 individuals)

-0.46



7-10 year olds
-0.52

-1.99



The data were then grouped by formation and resorption methods to analyze any
possible differences created by the two groups. As was stated earlier, it is important to
interpret the following statistics cautiously due to their relatively small sample sizes. The
analysis began with a paired t-test between the formation ages and the chronological
ages. The null and alternative hypotheses were determined to be the same as those stated
above for the overall sample. The means calculated by this test were 7.07 for the
formation group and 7.86 for the chronological age (Appendix Table 7). The t-value was
4.53 with 44 degrees of freedom and a two-tailed p value of .00 which can also be
interpreted as less than .01 (Appendix Table 8). These results indicate that the null
hypothesis should be rejected and it should be concluded that the age determination
technique based on the formation of teeth outlined by Moorrees et al. is inappropriate for
use on children with cerebral palsy.
A paired t-test was then performed on the chronological ages and the resorption
ages. The null hypothesis and the alternative hypothesis were determined once again to
be the same as those stated for the overall sample. The means of the chronological ages
and the resorption ages were calculated at 10.72 and 9.36 respectively (Appendix Table
9). The t-value was 2.47 with 8 degrees of freedom and a two-tailed p-value of .039
(Appendix Table 10). Because the p-value in this case was >.01, the null hypothesis was
accepted and it was concluded that there was not a statistically significant difference
between the mean of the chronological ages and the resorption ages. The implications of
this finding will be discussed further in the following section of this paper.
To compare the performance of the formation and resorption techniques, an
independent sample t-test was run testing the difference between the average estimated





ages and chronological ages (estimated age minus chronological age) grouped by
formation and resorption. The means calculated by this test were -1.36 for the resorption
group and -.79 for the formation group (Appendix Table 11). These results showed that
the resorption technique under-aged the sample to a larger degree than the formation
technique did. However, the two-tailed significance was .224 (Appendix Table 12),
which indicates that there was not a statistically significant difference between the aging
of both techniques.
Next, the data was grouped by sex. A paired t-test was conducted to determine
the difference between the chronological age and the estimated age of the male group.
The null hypothesis was that there was no difference between the means and the
alternative hypothesis was that the means differed, or in other words, the Moorrees et al.
method was under- or over-aging the male group. The chronological age mean was
found to be 8.55 and the mean of the estimated age for males was 7.61 (Appendix Table
13). The t value was 4.65 with 36 degrees of freedom and a p-value of .000 which can
also be interpreted as less than .01 (Appendix Table 14). The resulting means for both
groups demonstrated that the male group was being under aged and the null hypothesis
was rejected in this case due to the p-value being less than .01.
A paired t-test was then conducted to determine the difference between the
chronological age and the estimated age of the female group. The null and alternative
hypotheses were determined to be the same as those outlined for the male group. The
means determined by this test were 7.52 for the chronological age and 6.96 for the
estimated ages of the female group (Appendix Table 15). The t value was 1.88 with 13
degrees of freedom and a two-tailed p-value of .083 (Appendix Table 16). Unlike the





male group, the p-value in this case was greater than .01 allowing the null hypothesis to
be accepted.
The last test to be performed was an independent sample t-test to determine if
there was a significant difference between the aging of males and females in this study.
The difference between chronological age and estimated age was defined as estimated
age minus chronological age and the resulting number was grouped by sex. The
independent sample t-test resulted in mean values of -.91 for the males and -.56 for the
females (Appendix Table 17. The two-tailed p-value was .352, which indicates that there
was not a statistically significant difference between the aging of the males and females
(Appendix Table 18).

DISCUSSION
The results of this study showed that there was a significant positive correlation
between the estimated ages and the chronological ages, demonstrating the clear link
between chronological age and the dental development of children with cerebral palsy.
This positive correlation is evident graphically (Appendix Figure 6) with the data evenly
dispersed around and following the regression line. Also evident in this graph is the
absence of any significant outliers. Although this graph demonstrates the presence of a
correlation between the chronological ages and dental development, it does not speak in
any way to the accuracy of the age estimation that resulted from the use of the Moorrees
et al. method or the rate at which the dental development of children with CP occurs.
To determine this, an independent t-test was conducted and the results showed
that there was a statistically significant difference between the means of the chronological





ages and the estimated ages. This test clearly demonstrated that the Moorrees et al.
method of age estimation was unable to accurately estimate the ages of the individuals
with cerebral palsy, resulting in the conclusion that the dental development of these
individuals falls outside of the norms created for the unaffected children used in the
Moorrees et al. (1963 a,b) sample. More specifically, a comparison of the means
between the chronological age and the estimated age showed that the Moorrees et al.
method under-aged 75% of the study sample by an average of 1.25 years (Appendix
Figure 7). Moreover this research demonstrates that some individuals were under-aged
by as many as 4.5 years.
To gain a better understanding of the degree of error in age, the sample was
divided into three age groups: 3-6 year olds, 7-10 year olds, and 11-15 year olds. The
average error in estimation for the first two groups were -.46 years and -.52 years
respectively. Although many of the individual estimates technically still fit into the two
standard deviations that the Moorrees et al. method allows in age determination, the
average error calculated in this study represents a negative shift in the average of the
dental development of children with cerebral palsy. Even though half of a year may seem
like a short period of time, it can have a large impact when trying to estimate the age of
individuals in these younger categories.
The final age category, the 11 to 15 year olds, experienced the greatest amount of
error in estimation with an average of -1.99 years. As was the case with the younger
categories, most of these age estimates in the older category would technically fall into
the two standard deviations allowed by the Moorrees et al. method; however, this average
error represents a negative shift in the dental development of this part of the sample. This





shift of almost two years is a negative shift in the average age of the sample,
demonstrating a clear delay in the dental development of the children with cerebral palsy
used in this study.
The sample was then broken down into smaller groups to gain a more fine-tuned
understanding of the composition of the sample. Paired t-tests were run between the
chronological ages and the ages determined using both the Moorrees et al. formation and
resorption techniques in order to compare the performance of both. The results of the
paired t-test showed that the means of the formation ages and the chronological ages were
different to a significant degree, resulting in the conclusion that the norms created for the
Moorrees et al. sample based on dental formation are not applicable to the sample of
children with cerebral palsy and should not be applied. From this, it can also be
concluded that the rates of dental formation of children with cerebral palsy are different
than unaffected individuals. However, the paired t-test that was run between the
resorption ages and the chronological ages showed that there was not a statistically
significant difference between the means of both, allowing the assumption that the rates
of resorption between the individuals with cerebral palsy and unaffected individuals is
similar.
This particular finding is interesting and somewhat contrary to what is found in
other studies. Megyesi et al. (2009), for example, described a case in which an individual
with cerebral palsy still retained several deciduous teeth even in his early 20’s. Staufer et
al. (2009), also describes two case studies where the deciduous teeth failed to resorb and
had to be extracted to allow for the eruption of the permanent dentition. The common
factor that these three studies, and others like it share, is the fact that the individuals being





studied had all been diagnosed with severe forms of cerebral palsy. This issue is
important for a couple of reasons. To begin with, severe bruxism often accompanies
severe forms of cerebral palsy due to the increasing loss of control of the muscles of the
face as the severity increases. In the most severe forms of CP, the bruxism is generally
also extremely severe. This can result in the complete failure of the deciduous dentition
to exfoliate, leaving both sets of dentition somewhat intact.
The severity factor is also important due to the fact that most dental professionals
find it hard to medically treat individuals with severe forms of cerebral palsy. This is in
most part due to the fact that these individuals generally cannot sit for long periods of
time in examination chairs or in positions that are most conducive to examination and
documentation, including the taking of radiographs. Very often, the dental treatment
plans of individuals with severe cerebral palsy will not include dental x-rays simply
because the individual is unable to hold in a position that would make radiography
possible (Miller 2005). This latter point is important to the current study due to the fact
that there is a likely potential for individuals with severe forms of cerebral palsy to be
excluded or poorly represented in studies that use dental radiographs due to the fact that
radiographs of their dentition are so uncommon.
The last test to be performed on this category was an independent samples t-test to
compare the performance of the formation and resorption methods. The results of the
independent t-test showed that there was not a significant difference between the
performances of either method. It should be noted, however, that there is a very large
difference between the sample sizes used in this test. Out of the entire study sample of 51
individuals, 45 of those individuals had radiographs that were suitable for testing with the





formation method while only 9 individuals had radiographs that were suitable for testing
with the resorption method. As a result, the makeup of these two smaller samples was
also very different. For example, the average chronological age of the formation group
was 7.86 years with an estimated age of 7.07 years (Appendix Table 7). The resorption
group, on the other hand, had an average chronological age of 10.72 years and an
estimated age of 9.36 years (Appendix Table 9) making it on average significantly older
than the formation group. These results, along with those produced by the paired t-test
on the resorption group, should be interpreted carefully due to the extremely small
sample size. Small samples such as these are extremely susceptible to sampling error
which may mask what is really occurring within the sample (Moore et al. 2003).
The sample was then broken down into male and female categories to better
understand the sex differences in dental developments. To begin with, a paired t-test was
performed to compare the means of the chronological ages with the estimated ages of
males and females. The results showed that there was indeed a statistically significant
difference between the means of the chronological ages and the estimated ages of males.
These results also show that, like the overall sample, the Moorrees et al. method of age
estimation underestimated the sample of males in the current study. This leads to the
conclusion that the dental development of the males in the current study is delayed
relative to the sample used in the Moorrees et al. study. The results of the paired t-test on
the female group, however, were found to be not statistically significant when the means
of the chronological ages and the estimated ages were compared. Once again, however,
it is important to mention the sample sizes used for these tests, which were 37 individuals
for the male group and only 14 individuals in the female group. As was with the tests on





the resorption group, these results must be interpreted carefully due to the potential
affects of sampling error on such small sample sizes.
The last test to be run was an independent t-test between the male and female
groups with the purpose of better understanding if a difference existed between the aging
of males and females in this study. The results of the test showed that males seemed to
be under-aged to a larger degree than females; however, there was not a statistically
significant difference between the aging of both groups. Once again it is important to
note the small sample sizes used for this test and the potential for sampling error.
This issue of sample size is an important one to consider. It is often difficult to
determine an adequate number for the size of a sample in a given study and in the case of
this thesis, also difficult to obtain enough data to make up that sample. Statistics are
based on probability and are highly affected by the variability that exists within the
sample. As a result, the amount of error that occurs when calculating statistics will
decrease as the size of the sample increases. In other words, larger samples experience a
significantly lower amount of error. Large samples are ideal, however not always
attainable. Unfortunately, the minimum number of individuals that make up an
acceptable sample is highly debated within the scientific community. While some
consider anything more than 30 to be a large sample, there are those who believe samples
should always be larger than 100 or even 200 (Cohen, 1990; Sokal and Rohlf, 1995;
Hogg and Tannis, 2008). The sample used in this thesis totaled 51 individuals and was
broken down into groups as small as 9 individuals. Although the statistical tests run on
the larger samples may not have been affected by error to a large degree, the tests run on





the smaller samples should be interpreted with a great amount of caution due to a high
possibility of error influencing the results.
In addition to the error caused by a small sample size, the scientific literature also
suggests that there could be other issues that might be influencing the results of this
thesis. The first influencing factor worth considering is the possibility of co-morbidities.
It is very rare that CP acts in isolation to all other developmental disabilities. Because CP
results from severe trauma to the brain, it is highly likely that children with CP will also
present with other disorders such as mental retardation, hearing and vision disabilities,
and epilepsy (Low et al. 1982; Weddell et al., 2004). It is unknown how these other
disabilities work in conjunction with CP on the development of the human body and what
their specific individual contributions to the development of the dentition truly are.
One of the major benefits of using the dentition in age estimation is its apparent
ability to be buffered against environmental factors to a larger degree than bone is
(Ubelaker, 1986). However, the affects of these environmental factors are still important
to consider when they are exaggerated in any way. Nutritional deficiency is one such
environmental factor that can be exaggerated when a severe form of CP causes great
difficulties in the feeding abilities of children. These difficulties can range from having a
hard time chewing, due to poor muscle control, to the inability to eat without the use of a
feeding tube (Gisel and Patrick, 1988; Fang et al. 2002). As a result, these children
generally end up with varying magnitudes of malnutrition which in turn leads to what is
known as growth failure, or depleted subcutaneous fat and muscle stores. Interestingly,
the magnitude of the severity of malnutrition and its affect on growth in children has been
found to be linked with the magnitude of the severity of the CP (Stallings et al., 1993a;





Stallings et al., 1993b). In other words, as the severity of the CP increases, so does the
severity of the malnutrition and the resulting growth failure. Unfortunately, the affects of
nutrition on children with CP have only been studied in regards to the growth of the body
instead of the development of the dentition. As a result, very little is still known about
how nutrition directly affects the dentition in children with CP.
The final factor that could be affecting the results of this thesis is also the most
complicated and one that has already been briefly discussed in this thesis. It is what is
known as disease severity variables and includes all of the CP symptoms that are a direct
result of the original brain trauma (Stevenson et al. 1995). As stated earlier, the severity
of the original trauma to the brain is proportionate to the severity of the resulting
symptoms, or in other words, as the severity of the trauma increases, so does the severity
of the symptoms of CP. It has been shown that as the severity of both the trauma and the
symptoms increase, so too does the rate of growth failure in the form described above,
but also in limb length, limb circumference, and joint breadth (Stevenson et al., 1995).
The mechanism by which this works is unknown, however several hypotheses have been
made to explain this phenomenon which include the disuse of muscles, reduced blood
flow, and sensory deficits. Whatever the explanation may be, cerebral palsy itself has
been shown to cause a delay in the growth and development of the body in differing
degrees of severity. This continuum of severity most likely acts on the development of
the dentition in a similar way. As the severity of the CP increases, so too does the
severity of the mechanisms working on the dentition to delay development, most likely in
varying degrees as well.





What is left is a rather complex picture of how trauma to the brain early in life can
affect the development of the human body and dentition. Although it has been shown
both in the scientific literature and in this thesis that cerebral palsy can cause a delay in
the growth of the body and the dentition, an explanation of the exact mechanisms by
which this works is still missing. Further research is needed to better understand and
separate out the affects of each piece of the complex puzzle that surrounds a diagnosis of
cerebral palsy. Most importantly, a large sample size would benefit this research greatly
due to the fact that it could encompass more of this variability and provide a more
accurate picture of the complex nature of CP and its affects on development. A larger
sample size is also needed to attain a more accurate view of the development of the
dentition of females, as well as root resorption in general, due to the extremely small
sample sizes used in study.
The purpose of this thesis was to examine the effects of cerebral palsy on the
development of the dentition and how those effects can influence the reliability and
accuracy of the Moorrees et al. method of age estimation. The specific aims were
twofold: (a) to determine the correlation between chronological age and dental age
among children with cerebral palsy and (b), to determine the reliability and accuracy of
the Moorrees et al. method of assessing age using tooth formation and root resorption on
individuals with cerebral palsy. Based on the results of this preliminary analysis, the
conclusion that was reached was that the sample of children with cerebral palsy used in
this study did indeed show a delay in dental development relative to the sample of
unaffected children used by Moorrees et al. As a result, the Moorrees et al. method of
age estimation was found to be inappropriate for use on the sample of children with





cerebral palsy. Furthermore, additional research is needed to establish the exact
mechanisms that cause this delay, as well as how the varying degrees of severity of
cerebral palsy can change the rates of development of the dentition.

CONCLUSION
Despite advancements in diagnosis and treatment, cerebral palsy remains a
common condition in industrialized nations with 2 out of every 1000 live births affected
(Panteth and Kiel, 1984). In addition to the relative high frequency of CP, the Westat
Corporation found that children with disabilities are twice as likely to be the victims of
neglect and abuse (Westat, 1993). With these two powerful statistics in mind, it is highly
likely that a forensic anthropologist will become involved in a case where cerebral palsy
is an issue. In the event that this does occur, an accurate estimation of age is highly
dependent on a complete understanding of how CP affects the use of traditional methods
of age estimation. More specifically, the research completed in this thesis sought to fill
in the gaps of what is known about the dental development of children with cerebral
palsy.
To accomplish this goal, a commonly practiced anthropological method of age
estimation was used to estimate the ages of a sample of children with cerebral palsy. The
Moorrees, Fanning, and Hunt (1963a,b) method of age estimation was used in this study
to estimate the ages of a sample of 51 children with a diagnosis of cerebral palsy from the
Pediatric Dentistry Service at the University of Michigan Health System Hospital
(UMHS) Dentistry Clinic. The sample was comprised of panoramic dental radiographs
and periapical radiographs from 37 males and 14 females between the ages of 3 and 15.





After the ages were estimated using the Moorrees et al. method, paired t-tests
were run to compare the means of the chronological age with estimated ages, male and
female ages, and the resorption and formation ages. Independent t-tests were also used in
this study to compare the difference between the female/male subgroups and the
resorption/formation subgroups.
In the beginning it was hypothesized that the dental development of the sample of
children with cerebral palsy would be significantly delayed to the point that the Moorrees
et al. method of age estimation would be found to be inappropriate for use on a study
sample of individuals with cerebral palsy. The results of the independent t-test
comparing the chronological ages with the overall estimated ages of the sample used in
this study supported this hypothesis by demonstrating that there was a significant
difference between the means of both groups. To be more specific, the Moorrees et al.
method under-aged 75% of the sample by an average of 1.25 years in this study.
When the sample was broken down into subgroups, it was found that there
seemed to be no difference between the performance of the formation and resorption
methods or the aging of males and females. Also, like the sample as a whole, both the
formation ages and the male ages were found to be significantly different than the
chronological ages. The resorption and female ages, however, were found to not be
significantly different from the chronological ages. Although it might be tempting to
assume that the dental development of individuals with cerebral palsy in general may fit
the norms created by Moorrees et al. in these last two categories, the extremely small size
of the samples used in these last two groups should be considered first.





Small samples are problematic for multiple reasons in this case. First of all, they
increase the likelihood for error. Secondly, because of the variability that exists within
cerebral palsy, it is likely that a certain type or severity of CP may have been excluded
from the test. With such a small sample size, it is difficult to pick up on the things that
could be having a more nuanced affect on development such as type of CP or comorbidities. This study would benefit greatly by the addition of a large sample where
more of this information is known and accounted for. Possibly then can the complexity
of the relationship between cerebral palsy and dental development be better understood.
Beyond this relationship lies the issue of how a change in the rate of dental
development affects the accuracy of traditional anthropological means of age estimation.
The estimation of age is a critical component of the biological profile and a successful
positive identification can sometimes hinge on this single component. As a result,
anthropologists like all scientists, are responsible for the continuous exercise of validating
the methods they use. This includes the discovery and subsequent study of anything that
could be affecting the accuracy of any method. It has been shown in several studies that
the norms created for the Moorrees et al. method are not applicable to various groups of
people due to natural human variation in dental development. The preliminary findings
of this thesis demonstrate yet another group that may not fit into the Moorrees et al.
norms for dental development. As a result, anthropologists should use extreme caution
when attempting to use this method, or any method based on dental development, on an
individual with cerebral palsy.
The purpose of this study was to begin the conversation about the effects of
cerebral palsy on dental development and how they can affect the traditional





anthropological methods of age estimation. The results of this study showed that the
dental development of the sample of children with cerebral palsy is delayed to a
significant enough degree to affect the accuracy of the Moorrees et al. method of age
estimation. Based on the results of this preliminary analysis, the use of the Moorrees et
al. age estimation technique on individuals with cerebral palsy is not recommended.
Further research is needed to better understand the relationship between cerebral palsy
and dental development and how other factors such as co-morbidities, severity, and type
of CP could be affecting the rates of development differently.





APPENDIX





Figure 1 – Stages of Deciduous Tooth Formation



















Figure 2 – Tooth Formation Stages and Symbols





Figure 3 – Example Graph of Morrees et al. Norms Used to Estimate Age







Figure 4 – Sample Radiograph with Moorrees et al. Stages





Table 4 – Pearson’s r Correlation Between Chronological Age (CA) and Estimated Age
(EA)
Correlations
CA
CA

Pearson Correlation

EA

1

.947*

Sig. (2-tailed)
N

51

EA Pearson Correlation
Sig. (2-tailed)
N

.000
51

.947*

1

.000
51

* Correlation is significant at the 0.01 level (2-tailed).
Figure 5 – Correlation Between Chronological Age and Estimated Age





51

Table 5– Paired Sample t-Test Statistics for Chronological Age (CA) and Estimated Age
(EA)
Paired Sample Statistics
Standard

Standard Error

Mean

N

Deviation

Mean

CA

8.26471

51

3.537800

.495391

EA

7.43373

51

2.969933

.415874

Table 6 – Paired Sample t-Test Results for Chronological Age (CA) and Estimated Age
(EA)
Paired Sample t-Test
t
CA - EA



df

Sig. (2-tailed)

4.970

50

.000



Figure 6 – Estimated Error: Chronological Age and Estimated Age minus Chronological
Age

Table 7 – Paired Sample t-Test Statistics for Chronological Age (CA) and Formation Age
(FA)
Paired Sample Statistics
Standard

Standard Error

Mean

Deviation

Mean

CA

7.8611

45

3.51117

.52341

FA



N

7.0667

45

3.01868

.45000



Table 8 – Paired Sample t-Test Results for Chronological Age (CA) and Formation Age
(FA)
Paired Sample t-Test
t

Sig. (2-tailed)

4.530

CA - FA

df
44

.000

Table 9 – Paired Sample t-test Statistics for Chronological Age (CA) and Resorption Age
(RA)
Paired Sample Statistics
Standard

Standard Error

Mean

N

Deviation

Mean

CA

10.7222

9

2.05945

.68648

RA

9.3611

9

1.06881

.35627

Table 10 – Paired Sample t-Test results for Chronological Age (CA) and Resorption Age
(RA)
Paired Sample t-Test
t
CA – RA



df

Sig. (2-tailed)

2.469

8

.039



Table 11 – Group Statistics for Independent Sample t-Test of the difference between the
Chronological Age (CA) and the Estimated Age (EA) when grouped by Formation
(Group 1) and Resorption (Group 2) Techniques
Groups Statistics For Independent Sample t-Test
Standard

Standard Error

N
Difference EA – CA

2

Deviation

Mean

45

1

Mean
-.7938

1.17652

.17539

9

-1.3611

1.65412

.55137

Table 12 – Results of the Independent Sample t-Test of the difference between the
Chronological Age (CA) and the Estimated Age (EA) when grouped by Formation
(Group 1) and Resorption (Group 2) Techniques
Independent Sample t-Test
t

Sig. (2-tailed)

-1.231

Difference EA - CA

df
52

.224

Table 13 – Paired Sample t-Test Statistics for the Chronological Age (CA) and Estimated
Age of Males (EAM)
Paired Sample Statistics
Standard

Standard Error

Mean

Deviation

Mean

CA

8.54730

37

3.613650

.594080

EAM



N

7.61365

37

3.118613

.512697



Table 14 – Paired sample t-Test results for Chronological Age (CA) and Estimated Age
of Males (EAM)
Paired Sample t-Test Results
t

Sig. (2-tailed)

4.647

CA - EAM

df
36

.000

Table 15 – Paired Sample t-Test Statistics for the Chronological Age (CA) and Estimated
Age of Females (EAF)
Paired Sample Statistics
Standard

Standard Error

Mean

N

Deviation

Mean

CA

7.5179

14

3.33897

.89238

EAF

6.9582

14

2.58003

.68954

Table 16 – Paired Sample t-Test results for Chronological Age (CA) and Estimated Age
of Females (EAF)
Paired Sample t-Test Results
t
CA - EAF



df

Sig. (2-tailed)

1.881

13

.083



Table 17 – Group Statistics for Independent Sample t-Test of the difference between the
Chronological Age (CA) and the Estimated Age (EA) when grouped by Males (Group 1)
and Females (Group 2)
Group Statistics for Independent Sample t-Test
N

Standard

Standard

Deviation
Difference EA – CA

Mean

Error Mean

1

37

-.9135

1.23012

.20223

2

14

-.5594

1.11369

.29765

Table 18 – Results of the Independent Sample t-Test between Chronological Age (CA)
and Estimated Age (EA) when grouped by Sex.
Independent Sample t-Test Results
t
Difference EA-CA



df

Sig. (2-tailed)

-.941

49

.352



BIBLIOGRAPHY





BIBLIOGRAPHY
Bass, WM. 2005. Human Osteology: A Laboratory and Filed Manual, 5th Edition.
Columbia MO: Missouri Archaeological Society.
Bhat M and Nelson K. 1989. Dental enamel defects in primary teeth in children with
cerebral palsy, mental retardation, or hearing defects: A review. Advances in Dental
Research, 1989(3):132-142.
Buikstra, J and Ubelaker D. 1994. Standards for Data Collection from Human Remains.
Fayetteville: Arkansas Archaeological Survey.
Cardosa, H. 2007. Environmental Effects on Skeletal Versus Dental Development: Using
a Documented Subadult Skeletal Sample to Test a Basic Assumption in Human
Osteological Research. American Journal of Physical Anthropology, 132:223-233.
Cohen J. 1990. Things I have learned (so far). Am Psychol 45(12):1304-1312.
Davidoff A, Winkler S, Lee M. 1972. Dentistry for the Special Patient: The Aged,
Chronically Ill, and Handicapped. Philadelphia, PA: WB Saunders Company.
Daubert v. Merrell Dow Pharmaceuticals. 1993. 509 U.S.C 579 as found at
http://caselaw.lp.findlaw.com/scripts/getcase.pl?court=US&vol=509&invol=579
Demirjian A. 1985. Interrelationships among measures of somatic, skeletal, dental and
sexual maturity. American Journal of Orthodontics 88:433-438.
Demirjian, A. 1986. Dentition, 2nd edition. In: Falkner F and Tanner JM, editors. Human
growth: A comprehensive treatise. New York: Plenum Press; 269-298.
Dixon, L. 2001. Changes in the standards for admitting expert evidence in federal civil
cases since the Daubert Decision. Santa Monica CA: Rand Corporation.
Fang L, Fung E, Sallings V et al. 2002. Relationship of nutritional status to health and
societal participation in children with cerebral palsy. The Journal of Pediatrics.
141(5):637-643.
Franklin, D, Luther F, Curzon, M. 1996. The prevalence of malocclusion in children
with cerebral palsy. European Journal of Orthodontics, 18:637-643.
Gisel E and Patrick J. 1988. Identification of children with cerebral palsy unable to
maintain a normal nutritional status. The Lancet 33(8580):283-286.
Gron AM. 1962. Prediction of tooth emergence. J Dent Res 41: 573-585.





Harris E, McKee J. 1990. Tooth mineralization standards for blacks and whites from the
middle southern united states. J Forensic Sci 35:859-872.
Hogg, RV and Tannis E. 2008. Probability and Statistical Inference. New York:
Pearson, Inc.
Iscan M and Loth S. 1986. Determination of age from the sternal rib in white females: A
test of the phase method. Journal of Forensic Science, 31(3): 990-999.
Iscan, MY and Loth, SR. 1989. Osteological manifestations of age in the adult. In:
Iscan MY and Kennedy KAR, editors. Reconstruction of Life from the Skeleton. New
York: Wiley-Liss, Inc; 23-40.
Keinan D, Smith P, Zilberman U. 2006. Microstructure and chemical composition of
primary teeth in children with Down syndrome and cerebral palsy. Archives of Oral
Biology, 51:836-843.
Krogman W. 1962. The Human Skeleton in Forensic Medicine. Springfield, IL: Charles
C Thomas Publisher.
Liversidge H, Herged B, Rosing F. 1998. Dental age estimation of non-adults. A review
of methods and principles. In: KW Alt, FW Rosing and M Teschler-Nicola, editors.
Dental Anthropology – An Introduction. Vienna: Springer.
Low NL and Downey JA. 1982. Cerebral Palsy. In: JA Downey and NL Low editors.
The Child with Disabling Illness: Principles of Rehabilitation. New York, NY; Chapter
4.
Megyesi MS, Tubbs RM, Sauer NJ. 2009. An analysis of human skeletal remains with
cerebral palsy: Associated skeletal age delay and dental pathologies. Journal of Forensic
Science, 54(2):270-274.
Miller F. 2005. Cerebral Palsy. New York: Springer.
Moller P. 1973. Treatment of the Handicapped Child. In: SB Fin editor. Clinical
Pedodontics. Philadelphia, PA: WB Saunders Company.
Moore DS, McCabe GP. 2003. Introduction to the practice of statistics. New York: WH
Freeman and Company.
Moorrees C, Fanning E, Hunt E. 1963a. Formation and resportion of three deciduous
teeth in children. American Journal of Physical Anthropology, 21:205-213.
Moorrees C, Fanning E, Hunt E. 1963b. Age variation of formation stages for ten
permanent teeth. J Dent Res 42(6):1490-1502.





Ozerovic B. 1980. Correlation of dental and skeletal age in children with cerebral palsy.
European Journal of Orthodontics, 2:193-195.
Paneth N and Kiely J. 1984. The frequency of cerebral palsy: a review of population
studies in industrialized nations since 1950. In: E. Alberman, F. Stanley (eds)
Epidemiology of the cerebral palsies. Spastics International Publications, London, pp.
46-56.
Pellegrino, L. 2002. Cerebral Palsy. In: Batshaw ML, editor. Children with
Disabilities. Baltimore MD: Paul H Brooks Pub. Co. Inc.
Philips VM, van Wyk Kotze TJ. 2009. Testing standard methods of dental age
estimation by Moorrees, Fanning and Hunt and Demirjian, Goldstein and Tanner on three
South African children samples. J Forensic Odontostomatol, 27(2): 20-28.
Pope J and Curzon M. 1991. The dental status of cerebral palsied children. Pediatric
Dentistry, 13(3): 156-162.
Rodrigues dos Santos MT, Masiero D, Novo NF, Simionato MRL. 2003. Oral
conditions in children with cerebral calsy. Journal of Dentistry for Children (Chicago),
70:40-46.
Rosenbaum CH, McDonald RE, Levitt EE. 1966. Occlusion of cerebral –palsied
children. Journal of Dental Research. 45: 1696-1700.
Ubelaker, DH, 1986. Estimating age at death from immature human skeletons: An
Overview. Journal of Forensic Sciences, 32(5):1245-1263.
Scheuer L, Black S. 2000. Developmental juvenile osteology. San Diego CA:
Academic Press.
Scheuer L, Black S. 2004. The juvenile skeleton. Amsterdam: Elsevier Academic Press.
Schour, I and Massler, M. 1941. Development of the human dentition. Journal of the
American Dental Association, 28:1153-1160.
Shaw L, Weatherhill S, Smith A. 1998. Journal of Dentistry for Children, 65(6): 484486.
Smith BH. 1991. Standards of human tooth formation and dental age assessment. In:
Advances in Dental Anthropology, Kelly and Larsen (eds). New York: Wiley-Liss.
Stallings V, Charney E, Daview J, Cronk C. 1993a. Nutrition-related growth failure in
children with quadriplegic cerebral palsy. Developmental Medicine and Child
Neurology. 35(11):126-138.





Stallings V,Charney E, Davies J, Cronk C. 1993b. Nutritional status and growth of
children with dipelgic or hemiplegic cerebral palsy. Developmental Medicine and Child
Neurology, 35(11)997-1006.
Stanley F, Blair E, Alberman E. 2000. Cerebral Palsies: Epidemiology and Causal
Pathways. London: MacKeith Press.
Staufer K, Hamadeh S, Gesch D. 2009. Failure of tooth eruption in two patients with
cerebral palsy and bruxism- a 10-year follow-up: a case report. Spec Care Dentist,
29(4):169-174.
Stewart TD, Kerley ER. 1979. Essentials of forensic anthropology especially as
developed in the United States. Springfield IL: Charles C Thomas Publisher.
Troutman KC, Full CA, Bystror FB. 1982. Developmental disabilities: Considerations
in dental management. In: RE Stewart, TX Barber, KC Troutman, SH Wei editors.
Pediatric Dentistry: Scientific Foundations and Clinical Practice. St. Louis, MO: Mosby
Inc.
Weddell JA, Sanders BJ, Jones JE. 2004. Dental problems of children with disabilities.
In: RE McDonald, DR Avery, and JA Dean editors. Dentistry for the Child and
Adolescent. St Louis, MO: Mosby Inc.
Westat Corporation. 1993. The incidence of maltreatment among children with
disabilities. Washington, DC: NCCAN.



