Download CONE BEAM COMPUTED TOMOGRAPHY THREE-DIMENSIONAL RECONSTRUCTION FOR EVALUATION OF THE MANDIBULAR CONDYLE

CONE BEAM COMPUTED TOMOGRAPHY THREE-DIMENSIONAL
RECONSTRUCTION FOR EVALUATION OF THE
MANDIBULAR CONDYLE
Brian Albert Schlueter, D.M.D.
An Abstract Presented to the Faculty of the Graduate School
of Saint Louis University in Partial Fulfillment
of the Requirements for the Degree of
Master of Science in Dentistry
2007
Abstract
Cone beam computed tomography (CBCT) has proven to
be a valuable imaging modality for examination of the
temporomandibular joint (TMJ).
CBCT three-dimensional (3D)
reconstructions potentially provide the clinician with
efficient and effective means for evaluation of the
mandibular condyle.
The purpose of the present study is to
investigate what the ideal Hounsfield unit values are for
the examination of the condyle; and if an ideal window
width and level can be identified, can one reliably
evaluate the mandibular condyle using the CBCT 3D
reconstruction?
Linear dimensions between six anatomical sights
were measured with a digital caliper to assess the anatomic
truth for 50 dry human mandibular condyles.
Condyles were
scanned with the i-CAT (Imaging Sciences International,
Hatfield, PA) CBCT; and image reconstruction and assessment
was accomplished using V-works™4.0 (Cybermed Inc, Seoul,
Korea).
Three linear three-dimensional measurements were
made on each of the 50 condyles at eight different
Hounsfield unit (HU) windows.
These measurements were
compared with the anatomic truth.
Volumetric measurements
were also completed on all 50 condyles, at 23 different
1
window levels, in order to define the volumetric
distribution of bone mineral density (BMD) within the
condyle.
Significance testing for 3D linear measurement
and volumetric differences was accomplished using
independent t-tests with a 95% confidence interval.
Significant differences were found in two of the
three linear measurement groups at and below the
recommended viewing window for osseous structures.
The
most accurate measurements were made within the soft tissue
range for HU window levels.
Volumetric distribution
measurements revealed that the condyles were mostly
comprised of low density bone; and that condyles exhibiting
significant changes in linear measurements were shown to
have higher percentages of low density bone than those
condyles with little change from the anatomic truth.
Assessment of the mandibular condyle, using the 3D
reconstruction, is most accurate when accomplished at
density levels below that recommended for osseous
examination.
Utilizing lower window levels, extending into
the soft tissue range, may compromise one’s capacity to
view the bony topography.
This would suggest that CBCT 3D
reconstructed images, by themselves, may not be a reliable
means for the diagnosis condylar pathology, and or changes
in condylar morphology.
2
3
CONE BEAM COMPUTED TOMOGRAPHY THREE-DIMENSIONAL
RECONSTRUCTION FOR EVALUATION OF THE
MANDIBULAR CONDYLE
Brian Albert Schlueter, D.M.D.
A Thesis Presented to the Faculty of the Graduate School
of Saint Louis University in Partial Fulfillment
of the Requirements for the Degree of
Master of Science in Dentistry
2007
COMMITTEE IN CHARGE OF CANDIDACY:
Assistant Professor Ki Beom Kim,
Chairperson and Advisor
Assistant Professor Donald R. Oliver,
Professor Gus G. Sotiropoulos
i
DEDICATION
I dedicate this thesis project to my wife Hollie and our
sons Noah and Parker.
ii
ACKNOWLEDGEMENTS
In acknowledgement of the people who have helped on
this thesis project.
I would like to thank Dr. Ki Beom Kim
for his constant support and dedication through the entire
process.
I would also like to thank Drs. Donald Oliver and
Gus Sotiropoulos for devoting their time, knowledge, and
encouragement along the way.
Also, this project would not
have been possible without the help of Dr. Cyrus Alizadeh,
Dr. Becky Schreiner and the Alizadeh Orthodontics staff.
Thank you for your gracious hospitality and selfless
contribution to help make this project a success. Finally,
thanks to Drs. Heidi Israel and Binh Tran for their help
with the statistical analysis of the data.
iii
TABLE OF CONTENTS
List of Tables............................................vi
List of Figures..........................................vii
CHAPTER 1: INTRODUCTION....................................1
CHAPTER 2: REVIEW OF THE LITERATURE
Development of Dental Radiographic Imaging......5
X-Ray .........................................5
Cephalometry ..................................6
Panoramic Radiography .........................7
Tomography ....................................9
Computed Tomography ..........................12
Cone Beam Computed Tomography ................17
Conventional CT vs. CBCT.......................19
Imaging Performance ..........................21
Radiation Dosimetry ..........................25
Size and Cost ................................28
CBCT Orthodontic Applications..................30
Accuracy .....................................30
Impacted Teeth and Oral Abnormalities ........34
Airway Analysis ..............................38
Orthognathic Surgery .........................40
Temporomandibular Joint Evaluation: Imaging...41
2D Imaging ...................................43
3D Imaging ...................................47
Hounsfield Units as a Measure of Bone Density
Distribution .................................50
References.....................................62
CHAPTER 3: JOURNAL ARTICLE................................70
Abstract.......................................70
Introduction...................................72
Materials and Methods..........................76
Selecting the Sample .........................76
Imaging ......................................77
Isolating and Measuring the Condyle ..........79
Volume Measurements at Varying Window Widths .86
Data analysis ................................89
Results........................................91
Discussion.....................................95
Conclusions...................................102
Acknowledgements..............................104
Literature Cited..............................105
iv
Vita Auctoris............................................109
v
List of Tables
Table 2.1:
Window widths for density values and
Hounsfield numbers in the body............61
Table 3.1:
Hounsfield Unit window widths (w) used to
create 3D renderings of the condyles. 3D
linear measurements were accomplished on
each of these
renderings................................80
Table 3.2:
Definitions of condylar linear
measurements..............................82
Table 3.3:
Definitions of anatomic landmarks.........82
Table 3.4:
Hounsfield Unit window widths (WW) used to
create 3D condyle renderings for volumetric
evaluation. Window 24 is equal to the sum
of window widths 1-23.....................89
Table 3.5:
Hounsfield Unit window widths (WW) used to
create 3D condyle renderings for volumetric
measurements (see Table 3.4). CL1
represents the average % volume of the
condyles that showed the greatest change in
condylar length measurement at 176 HU–2476
HU. CL0 was the least change in condylar
length measurement. CW1 was the greatest
change in condylar width, and CW0 the
least.....................................91
Table 3.6:
Hounsfield Unit window widths (W) used to
create 3D condyle renderings for linear
measurements (see Table 3.1 for W ranges).
Average condylar length (CL), condylar Width
(CW), and condylar height (CH) are compared
with the gold standard (GS)...............93
Table 3.7:
Comparison (Independent t-tests) of
volumetric measurements: CL0 vs. CL1, and
CW0 vs. CW1 at twenty-three window
levels....................................94
Table 3.8:
Window widths for density values and
Hounsfield numbers in the body............99
vi
List of Figures
Figure 1.1:
A multiplanar reconstruction (MPR) with
axial, coronal, and sagittal crossectional
views, as well as 3D reconstruction view...3
Figure 2.1:
Early laboratory CT machine, with x-ray
tube, designed by Godfrey N. Hounsfield
(image adopted and modified from Godfrey N.
Hounsfield’s Nobel lecture, 8 December,
1979).....................................14
Figure 2.2:
First clinical CT scanner prototype
installed at Atkinson Morley Hospital in
London, 1971 (image adopted and modified
from Godfrey N. Hounsfield’s Nobel lecture,
8 December, 1979).........................14
Figure 2.3:
Graphic representation of the image
acquisition technique the conventional CT
fan shaped beam and the CBCT cone shaped
beam......................................19
Figure 2.4:
Images adopted and modified from 3D
Diagnostix Inc. and IMTEC Imaging.........20
Figure 2.5:
Scale used by Hounsfield to demonstrate the
accuracy to which absorption values can be
ascertained on the CT picture.............52
Figure 2.6:
Hounsfield Scale of CT Numbers (image taken
from Jackson).............................53
Figure 2.7:
3D reconstruction (197 HU to 2476 HU).....58
Figure 2.8:
3D reconstruction (-100 HU to 2476 HU)....58
Figure 2.9:
3D reconstruction (-200 HU to 2476 HU)....58
Figure 2.10:
3D reconstruction (-400 HU to 2476 HU)....59
Figure 2.11:
3D reconstruction (-600 HU to 2476 HU)....59
vii
Figure 2.12:
These photographs of the mandible shown in
the previous reconstruction images
demonstrate the true appearance of the
articular surfaces of the right condyle. a)
lateral pole, b) posterior surface, c) right
mandible, d) posterior/lingual view right
half of the mandible......................60
Figure 3.1:
A multiplanar reconstruction (MPR) with
axial, coronal, and sagittal crossectional
views, and 3D reconstruction..............74
Figure 3.2:
Anatomical landmarks: a) anterior view of
condyle showing medial mandibular condyle
(MCo) and lateral mandibular condyle (LCo),
b) lateral view of condyle showing posterior
mandibular condyle (PCo), anterior
mandibular condyle (ACo), and superior
mandibular condyle (SCo), c) lingual view of
the ramus showing lingula (L).............77
Figure 3.3:
3D Reconstruction isolation: a) Initial
lateral view 3D reconstruction, b) Frankfort
Horizontal initial sculpting cut, c)
Vertical sculpting cuts, d) Completed
isolation for condylar measurements.......81
Figure 3.4:
Point-to-point linear measurements on the 3D
reconstruction in V-works™4.0 SSD mode; a)
Condylar width (CW), b) Condylar length
(CL), and c) Condylar height..............83
Figure 3.5:
Lateral view of plane construction........84
Figure 3.6:
Superior view of plane construction.......85
Figure 3.7:
Anterior view of plane construction with
(CW) measurement..........................85
Figure 3.8:
3D reconstruction isolation for volumetric
measurements of the mandibular condyle: a)
initial cut parallel to Frankfort
Horizontal, b) second cut parallel to the
first cut at the level of the most inferior
point in the sigmoid notch, c & d) lateral
and oblique views of the final isolated
reconstruction............................87
viii
Figure 3.9:
Anterior and lateral volumetric renderings
of the right condyle at four different
window widths: a) 176-2476 HU, b) 376-475
HU, c) 1476-1575 HU, d) 1976-2075 HU......88
Figure 3.10:
Distribution of % Differences between 3D
linear measurements and gold standards (GS);
a) CW measurement % differences in W8 (1762476 HU) with mean group (yellow) between
the red lines, b) CL measurement differences
in W8 with mean group (yellow) between the
red lines.................................92
Figure 3.11:
Distributions of condylar volume in twentythree window widths for the condyles of
groups CL1 and CL0, Window 1 represents the
lowest density bone, and window 23 being the
highest density bone observed.............96
Figure 3.12:
Distributions of condylar volume in twentythree window widths for the condyles of
groups CW1 and CW0, Window 1 represents the
lowest density bone, and window 23 being the
highest density bone observed.............97
ix
CHAPTER 1:
INTRODUCTION
Temporomandibular joint (TMJ) standard radiographic
studies such as the plain film radiography and panoramic
radiography have little capacity to reveal anything more
than gross osseous changes1 within the joint; and therefore,
in some cases a more comprehensive radiographic study is
indicated.
Radiographic analysis of the TMJ is a broad
field and is considered by some to be a separate subset of
oral and maxillofacial radiology,2 consisting of both two
and three-dimensional imaging modalities.
Two-dimensional
(2D) imaging of the TMJ employs conventional radiology to
produce a variety of projections.
The submental-vertex,
lateral transcranial, transpharyneal, transmaxillary (AP),
as well as conventional tomography are a few examples of
the 2D studies used in TMJ evaluation.
Three-dimensional
evaluations, such as computed tomography (CT) and magnetic
resonance imaging (MRI), have been utilized to some degree;
however, historically, high cost,3,4 large radiation
dosage,5,6 large space requirements,3,4 and the high level of
skill required for interpretation have kept its use to a
minimum.
With the introduction of limited cone-beam
technology, such deterrents of CT imaging have been greatly
1
diminished.
With five different cone-beam computed
tomography (CBCT) scanners now available on the world
market, lower radiation dosages,7-10 and lower costs,3 3D
radiography is likely to become more commonplace in the
dental profession.
Demonstrating a broad spectrum of
applications and vastly improved accuracy over 2D
radiography,4,11,12 CBCT proves to be an invaluable diagnostic
tool for the evaluation of the osseous structures of the
TMJ.
Three-dimensional imaging facilitates comprehensive
examination of the subarticular osseous surfaces of the TMJ
through two different viewing modalities.
One allows for
the tissue of interest to be examined through progressive
crossectional slices in a chosen plane of orientation;
while the other produces 3D reconstructions capable of
being manipulated in any given direction, thereby
facilitating and expediting the examination procedure.
Often, these two viewing methods are combined into what is
referred to as the multiplanar reconstruction (MPR).
MPR
viewing mode enables the examiner to simultaneously assess
crossectional and three-dimensionally reconstructed images
(see Figure 1.1).
Within the condyle there is variation in bone
density and composition.
Cortical bone, trabeculae, and
2
Figure 1.1. A multiplanar reconstruction (MPR) with axial,
coronal, and sagittal crossectional views, as well as 3D
reconstruction view.
intertrabecular tissues have varying densities and
mechanical properties.13-17
These differences present a
challenge when examining the bony subarticular surfaces of
the condyle with 3D CT imaging.
For computed tomography
density is often expressed in the form of CT Numbers or
Hounsfield Units.
From radiation intensity readings
acquired in the scanning process, the density or
attenuation values of tissues of interest can be
calculated.
The attenuation coefficients are normalized
with respect to water as the reference material, and a
magnifying constant is then applied to produce a CT
number.18
Hounsfield19 originally described the Hounsfield
3
Unit (HU) as an absorption value; and he constructed a
scale to demonstrate the accuracy to which the absorption
values could be ascertained on a visual image.
For the
machine he described,19 the scale ranged from air (-1000) at
the bottom of the scale, to bone (1000) at the top of the
scale.
Each number represents a different shade of gray
within the spectrum.
The range of tones between black and
white seen in an image can be limited to a large or small
window within the scale.
This window can then be raised or
lowered depending upon the absorption value of the material
of interest19.
The examiner must be able to decide what
window level and width will most accurately represent the
anatomical truth of a tissue under examination.
It is
important for the clinician to be able to reliably detect
osseous abnormalities such as cortical erosion,
osteophytosis, osteoarthritis, and sclerosis.
The purpose
of the present study is to investigate what the ideal
Hounsfield unit values are for the examination of the
condyle; and if an ideal window width and level can be
identified, can one reliably evaluate the subarticular
anatomy of the mandibular condyle using the CBCT 3D
reconstruction?
4
CHAPTER 2:
REVIEW OF THE LITERATURE
Development of Dental Radiographic Imaging
X-Ray
X-rays were an accidental discovery made by Wilhelm
Conrad Roentgen on November 8, 1885.20
The discovery was
made while he was studying cathode rays in a high-voltage,
gaseous-discharge tube (Crookes tube).
He discovered that
a nearby barium-platinocyanide screen began to fluoresce
when the tube was in operation.
From this he decided that
the fluorescence was the result of an invisible form of
radiation with greater penetrating capacity than
ultraviolet rays.21
To identify this invisible form of
radiation Roentgen, coined the term “X-ray”.
“X”, being
the unknown variable in mathematic equations, was an
appropriate name for these rays with an unknown nature.22
The X-ray was first introduced to the world when Roentgen
published his first paper: “On a New Kind of Ray, A
Preliminary Communication” (1895).
Soon after, in 1901, he
was awarded the Nobel Prize for his revolutionary work with
X-rays.20
The discovery of X-rays had a profound effect on a
myriad of industries including, but not limited to, the
5
medical and dental professions.
The year following
Roentgen’s first publication, C. Edmund Kells became the
first American dentist to take dental radiographs of a
living subject.
He was also the first to exhibit the
dental X-ray apparatus at a
dental meeting; thereby
introducing this new diagnostic capability to the dental
profession.22,23
Cephalometry
Radiographic cephalometry was introduced by Pacini
in 1922 for anthropometric purposes.24
He made the process
somewhat reproducible by immobilizing the head and using
set source to object and object to film distances.
This
provided a means for visualizing the anatomy; however, it
gave no measurability, and the reproducibility was
questionable.
Just four years later an orthodontist named
B. Holly Broadbent adapted the roentgenographic craniometer
as a means to stabilize the heads of living subjects in a
fixed position for imaging. Previously, the craniometer or
craniostat, developed by Dr. T. Wingate Todd, was used
solely for holding skulls in a fixed position while precise
lateral and posteroanterior radiographs were made.25
Broadbent introduced his head holder for roentgenographic
studies of the living head26 in his 1931 paper: “A New X-Ray
6
Technique and Its Application to Orthodontia”
The nuance
of Broadbent’s roentgenographic cephalometry revolutionized
orthodontics; making it possible for the clinician to
obtain standardized and accurate measurements from lateral
and posteroanterior radiographs. These two-dimensional (2D)
images have been used to study craniofacial growth, facial
types, and relationships of the jaws; teeth, and soft
tissues, and the positioning and morphology of various
other bony structures.
They also aid in treatment
planning, forecasting future relationships, and evaluating
the results of growth and treatment.
Cephalometry is the
only practical imaging technique that allows the
investigator to quantitatively evaluate of the spatial
relationships between cranial and dental structures.2
For
this reason cephalometry has remained relatively unchanged
today.
Panoramic Radiography
Prior to the development of the panoramic
radiograph there was no practical way to image the jaws and
teeth in one film.
Panoramic radiography originated from
the need to accurately image the jaws in their entirety on
a single film.
Attempts made to accomplish this began in
1922 when A. F. Zulauf (USA) described and patented a
7
method whereby a narrow beam scanned the upper or lower jaw
to create a panoramic image.
He named this device “the
panoramic x-ray apparatus”.
Shortly after, in 1933, H.
Numata (Japan) constructed a device suitable for clinical
examinations and labeled his method “parabolic
radiography”.
These earlier forms of panoramic radiography
utilized intraoral film which resulted in various practical
problems.27
However, in 1949, Y. V. Paatero (Finland)
published papers describing the basic principles of
panoramic radiography using extraoral film.
Paatero worked
closely with an engineer by the name of T. Niemien, to
further develop the technique; and their close
collaboration continued until Paatero’s death in 1963.
Production of the first commercial equipment began in 1960
with the introduction of the Orthopantomograph.
Niemien
was the driving force behind its development and marketing;
and the manufacturer was established under the name Palomex
(Panoramic Layer Observing Machine for Export).
Development and technological advancement of the equipment
and technique continued, and in 1988 the Scanora device
(Orion Corporation/Soredex, Finland) was introduced.
It
was the first machine to combine panoramic radiography and
multidirectional tomography; thereby, providing the
clinician with the ability to acquire cross-sectional
8
images of the jaws.27
Panoramic radiography, however, has
many shortcomings related to the reliability and accuracy
of size, location, and form of the images created.
The
origin of these shortcomings lies in the design of the
technique.
The panoramic image is made by creating a focal
trough or region of focus within a generic jaw form and
size.
The most reliable images are obtained when the
subjects jaw form and size most closely approximate the
generic jaw form and size.
Anything outside this will
introduce error in the image produced.2
Tomography
Film based tomography (analog tomography) was
designed for the purpose of creating clearer images of
objects lying within a particular plane of interest.28
The
technique to produce a tomogram is very similar to that of
the panograph in that the X-ray tube moves in the opposite
direction of the film cassette on the other side of the
patient.
During the rotation only one section of tissue
remains in focus, while tissues above or below the section
are blurred by the motion of the tube and cassette.29
Though, the concept of tomography had been
described as early as 1914 by a Polish radiologist named
Mayer, the first true analog tomograms were not constructed
9
until the early 1930s.
In Vallebona’s 1930 paper he
described a tomographic x-ray unit in which the tube and
cassette remained stationary and the patient rotated
between them.
Three years later he constructed a unit in
which the patient remained stationary and the tube and
cassette rotated around.
This device produced images of
little diagnostic capability in that only the structures
along the axis of rotation remained in focus.
Around the
same time as Vallebona, the Dutch physician Ziedses des
Plantes, recognized today as the founder of modern analog
tomography, introduced linear, pluridirectional and
multisection tomography.
Pluridirectional tomographic motion yielded fewer
image artifacts than did the linear motion utilized in
earlier machines.
The sectional images produced with this
technique were recognized with many names including:
planigraphy, stratigraphy, laminagraphy, body select
radiography, and zonography.29
Tomography evolved as a
method to overcome undesirable superimpositions that are
difficult to eliminate in plane film radiography. One
example of this is presented with the complex angulations
required in the transmaxillary technique to avoid
superimposition of the mastoid process on the joint image.30
10
The first American tomographic unit was developed
by Andrews of the Cleveland Clinic in 1936.
Shortly after,
Keiffer and Moore constructed the tomographic laminagraph
at the Mallinckrodt Institute in St. Louis, MO. It wasn’t
until 1949 that a tomographic machine was made commercially
available; and was reproduced under the name “Polytome” by
Masslot.
Soon after, hypocycloidal motion was added to
this unit, and through marketing by Phillips, it became the
gold standard of analog tomography.29
This and other
complex movements such as circular and ellipsoid, found on
newer computer controlled machines, helped to minimize the
problems presented in earlier machines that used linear or
straight line movement patterns.
These more simplistic
movements tended to degrade the overall image sharpness.1,30
These images often appear streaked.
Streaks appear when
the long axis of a structure lying outside the focal plane
is oriented parallel to the movement of the tube.28,30
Modern, complex-motion tomographic units can be optimized
to image any selected region of the facial skeleton.2
They
provide sharper images, greater density uniformity,
consistent magnification, and increased dimensional
stability.28
Though quite versatile in their applications,
images produced with analog tomography are limited to one
plane or slice of the anatomy under study.
11
They have no
capability of combining slices to create an appearance of
three-dimensional (3D) form as in computed tomography.
Computed Tomography
Computed Tomography (CT) is essentially an advanced
form of tomography that utilizes computerized storage of
data from a series of thin tomographic sections taken from
multiple directions.
The exposures are recorded by an
array of sensors positioned on the opposite side of a
rotating gantry from the radiation source.30
The principle
of CT evolved from the work of an Austrian mathematician,
Radon, who demonstrated in 1917 that a 3D image of an
object could be reconstructed from an infinite number of
two-dimensional (2D) projections.
His work focused on
equations that described gravitational fields, and had
nothing to do with medical imaging.29
In the late 1960s and early 1970s several groups
were investigating tomographic imaging from projections as
a possible diagnostic tool, or potential aid in treatment
planning and radiation therapy.
Godfrey Hounsfield, an
engineer at the Central Research Laboratories of ElectroMusical Instruments (EMI) Ltd. in England, brought this
technology to a clinical reality in the early 1970s.
His
first experimental apparatus yielded reasonable images of a
12
symmetrical object; however, the time required for a scan
was excessive, taking up to 9 days for data acquisition and
another 2.5 hours for data processing.
Much of this was a
result of the low intensity of the radiation provided by
the Am-241 x-ray source used in this early apparatus.
Soon
after, Hounsfield replaced the Am-241 source with an x-ray
tube and reduced the data acquisition time from 9 days to 9
hours29 (Figure 2.1).
This was still too long to make
diagnostically adequate images of the abdomen due to the
constant motility that occurs in the gastrointestinal
track.
Instead, the first CT scanner Hounsfield developed
for humans focused on the brain as the target organ (Figure
2.2).
In October 1971 the first image was obtained at the
Atkinson Morley Hospital in London.
The scan, with a data
acquisition of 4.5 minutes, clearly revealed a frontal lobe
tumor in a 41 years old woman.29
There are two CT scanner configurations: one where
both the x-ray tube and the detector array revolve around
the patient, and the other where only the x-ray tube
rotates; and radiation detection is accomplished by the use
of a fixed circular array of detectors.28,30,31
The later
configuration often utilizes greater than one thousand
detectors; and newer generations often incorporate multiple
rows of detector rings; making it possible to acquire
13
Figure 2.1. Early laboratory CT machine, with x-ray tube,
designed by Godfrey N. Hounsfield (image adopted and
modified from Godfrey N. Hounsfield’s Nobel lecture, 8
December, 1979)19
Figure 2.2. First clinical CT scanner prototype installed
at Atkinson Morley Hospital in London, 197119 (image adopted
and modified from Godfrey N. Hounsfield’s Nobel lecture, 8
December, 1979).19
14
multiple slices per tube rotation.31
This type of CT
scanner incorporates slip ring technology which enables the
x-ray tube to rotate continuously around the subject in one
direction.
The path the tube makes around the subject is a
spiral or helical pattern; hence the name spiral or helical
CT.
This allows larger anatomical regions of the body to
be imaged during a single breath hold, thereby reducing the
possibility of movement artifacts.31
Since the introduction
of Hounsfield’s first prototype, there has been a gradual
evolution to five generations of such systems. Each system
is classified on the organization of the individual parts
of the device and the physical motion of the beam in data
acquisition.7
CT offers several distinct advantages over
conventional film tomography.
The first of these is its
ability to produce images completely devoid of
superimpositions from structures superficial or deep to the
area of interest.
Second, due to the high contrast
capabilities of CT images, differentiation between tissues
with density differences of less than 1% can be
distinguished.
Conversely, conventional film radiography
requires a 10% physical density difference for the examiner
to be able to distinguish between tissues.
The third of
these advantages is demonstrated through its ability to
15
produce images that can be viewed in the axial, coronal, or
sagittal planes from a single imaging procedure.28 These are
referred to as multiplanar reconstructions.
It is also
possible to construct 3D images of the scanned structures.
Three-dimensional reconstructions provide precise
and detailed information to aid in the study of the
craniofacial complex, as well as treatment planning in
maxillofacial surgery.32
They provide enhanced images that
enable the practitioner to more accurately and efficiently
evaluate the subject of study.
Three-dimensional images can be viewed on the
screen or processed into plastic models via milling or
stereolithographic biomodeling.2,32
Biomodels improve the
quality and precision of diagnostic measurements,
facilitate communication between specialists, and allow for
simulation of surgical procedures.32
These models are often
used for clinical purposes in maxillofacial surgery such
as: evaluation of craniofacial anomalies, surgery planning,
reconstruction of craniofacial defects, primary
reconstruction in craniomaxillofacial trauma surgery,
custom cranioplasty, and for accurate, preoperative
adaptation of reconstruction plates or osteosynthesis
devices.32
They can also be used aid in the construction of
maxillofacial prostheses. One example of this is
16
temporomandibular joint replacement prosthetics; which
often involves replacement of the condyle and in some cases
the glenoid fossa.
Cone Beam Computed Tomography
Computed tomography has proven to be quite helpful
for dental diagnosis, however, conventional helical-CT
units were not originally developed for this purpose.
The
problems in adapting helical-CT scans for dental use
include: high cost, large space requirement, long scanning
time, high radiation exposure, and low resolution in the
longitudinal direction compared with it’s relatively high
resolution in the axial direction.33-35
The last of these is
a result of the method by which longitudinal images are
produced, through summation of axial CT images.35
Each
axial slice is produced by one revolution of the fan shaped
beam of x-rays.
Then, the axial slices are stacked in
order to create a complete image of the object under study.
In 1997, the Department of Radiology in the Nihon
University School of Dentistry set out to resolve some of
shortcomings of conventional CT when they developed a
radiological unit using a new technology known as limited
cone beam computed tomography.
This new machine, the
Ortho-CT,36 was refined and improved; and in 2000 the
17
technology was transferred to the Morita Corporation as the
3DX multi-image micro-CT (3DX).33
The original prototype
was based on existing technology in which film was replaced
by an image intensifier;33 and radiation source was a coneshaped x-ray beam that rotated around the subject being
examined (see Figure 2.3).35
The 3DX machine, marketed by
Morita Corporation, has an exposure time of 17 seconds,
close to that of a panoramic exposure, and the radiation
dose is about 1/100 of the helical-CT.35
Many other
machines have been produced and marketed since the
introduction of the 3DX.
Generally, most use the same
technology which involves a cone-shaped x-ray beam and an
image intensifying sensor that rotate around the subject
under observation.
The i-CAT (Imaging Sciences
International, Hatfield, PA) and the Iluma (IMTEC imaging,
Ardmore, OK) CBCT systems, however, use amorphous silicon
flat panel image detectors capable of producing less image
noise than image intensifier tube/charge-coupled device
systems.37
Some of the CBCT acquisition systems now available
on the world market7 include: the NewTom 3G by Quantitative
Radiology, the i-CAT by Imaging Sciences, the CB MercuRay
by Hitachi Medical, the 3D Accuitomo by J Morita
Manufacturing Corporation, and the Iluma by IMTEC imaging.
18
Kau, Richmond, Palomo, and Hans review four of these
systems in their 2005 article entitled “Three-dimensional
cone beam computerized tomography in orthodontics”.
All
five systems are shown in Figure 2.4.7
Figure 2.3. Graphic representation of the image acquisition
technique the conventional CT fan shaped beam and the CBCT
cone shaped beam.7
Conventional CT vs. CBCT
Conventional CT imaging has been employed for
dental purposes prior to the development of limited cone
beam technology; however, its use has been limited.
Helical CT systems are large, expensive machines designed
primarily for full-body scanning at a high speed to
19
minimize movement artifacts.
Dental facilities are not
well-suited to house these machines; where cost is a
factor, space is limited and imaging requirements are
limited to the head.
Cone beam technology utilizes x-rays
Figure 2.4. Images adopted and modified from 3D Diagnostix
Inc38 and IMTEC Imaging.39
much more efficiently, requires far less electrical energy,
and allows for the use of smaller, and less expensive x-ray
components than fan-beam technology.3
With CBCT becoming
the standard for 3D oral and maxillofacial imaging over
conventional CT, it is important to recognize how these
systems are different when it comes to:
imaging
performance, size, cost, energy requirements, radiation
20
dosimetry, and the level of skill needed to operate the
systems.
Imaging Performance
A 2005 study by Holberg, Steinhäuser, Geis, and
Janson40 compared the image quality of fine dental
structures using both CBCT and conventional dental CT.
The
CBCT imaging was carried out using the DVT-9000 (an earlier
version of the New Tom 3G), and conventional CT imaging was
accomplished using the Light Speed Ultra manufactured by
General Electric Company.
Over 200 teeth were examined
with both systems; and image quality assessment was carried
out by three radiologically-experienced clinicians with a
minimum of five years experience in analyzing tomographic
slices of the craniofacial complex.
The image quality of
the axial slices through the periodontal ligament space in
the root area was examined.
The comparison between the two
systems was limited to axial slices which allows for the
production of high resolution images in conventional CT.33-35
The authors concluded that in contrast to dental CT, there
were little or no metal artifacts around fillings or
implants when CBCT was employed.
A single metal filling
can render an entire axial slice with conventional CT
useless.
They also found CBCT to be superior when it came
21
to the examination of major dental and skeletal structures
such as relation of teeth and visualization of skeletal
structures.
However, when it came to examining fine
structures like the periodontal ligament space, enameldentin interface, and the boundary of the pulp cavity, the
dental CT was superior.
Much of the difference in image quality of the axial
slices can be explained by the difference in acquisition
times.
There was a difference of 77.9 seconds between CBCT
and the dental CT.
With the longer acquisition time, CBCT
had a much greater chance of producing movement artifacts
that can affect the clarity of entire scan.
With dental
CT, only the individual 1.1 second slice, in which the
movement occurred, will be affected.
Many of the newer
CBCT machines have much shorter acquisition times ranging
from 10-40 seconds,7 therefore, likely to produce much fewer
movement artifacts than earlier machines.
Hashimoto et al. in 200636 also studied the imaging
performance of CBCT in comparison with helical CT.
For the
study they employed the Asteion Super 4 edition (Toshiba,
Tokyo, Japan), a medical four-row multidetector CT (MDCT);
and the 3DX CBCT.
In the United States and Europe the 3DX
is marketed as the 3D Accuitomo by Morita Company.
The
purpose of this study was to evaluate the image quality of
22
3DX.
To accomplish this they compared 3DX images with MDCT
images in order to determine how well bone and tooth
conditions could be reproduced.
The subject of the study
was a dried right maxillary bone cut into eight slices 2 mm
thick toward the zygomatico-plate in parallel with the
midline plane.
The slices were embedded in dental acrylic
resin at the level of the maxillary crowns; and returned to
their pre-cutting position prior to imaging.
Slices were
imaged directly parallel to the median sagittal plane
corresponding to the direction in which the maxillary bone
had been cut.
Acquisition time was 17 seconds for the 3DX,
and 0.75 seconds per slice for a total of 5.5 seconds for
the Asterion.
The examiners found image quality and bone slice
reproducibility to be far superior with the CBCT in
comparison to the MDCT.
Reproducibility of the slices was
determined through visualization of bone trabeculae,
enamel, dentin, pulp cavity, periodontal ligament space,
and lamina dura.
These structures were evaluated from
printed images produced from both machines, and compared
with the actual slices.
Hashimoto et al. in 2003 showed
similar results in an earlier study when they compared
images of a maxillary central incisor and mandibular first
molar made with the same imaging systems.
23
This evaluation
also focused on fine structures such as: cortical bone,
cancellous bone, enamel, dentin, pulp cavity, periodontal
ligament space, and lamina dura.
The images made with the
3DX were evaluated as significantly higher in quality than
the corresponding MDCT images.34
It is important, in the interest of the current
study, to review the reliability of CBCT in the examination
of the TMJ bony anatomy.
With helical CT being the
standard for evaluation of the TMJ bony components, it is
intuitive that it be used in a comparison to determine the
diagnostic reliability of CBCT.
Honda, Larheim, Maruhashi,
Matsumoto, and Iwai did just this in their 2006 article:
“Osseous abnormalities of the mandibular condyle:
diagnostic reliability of cone beam computed tomography
compared with helical computed tomography based on autopsy
material”.6
The purpose of this study was to compare the
reliability of the 3DX with helical CT in detecting osseous
abnormalities of the mandibular condyle; using macroscopic
observation as the gold standard.
The helical CT scanner
used was a Lemage SXE (GEYMS, Tokyo, Japan).
Twenty-one
autopsy specimens were mounted in acrylic holders and
oriented according to clinical positioning for exposures
with the 3DX and helical CT.
The 21 condyles were
independently assessed by two oral and maxillofacial
24
radiologists with 20 to 25 years experience.
The condyles
were assessed for osseous abnormalities such as: cortical
erosion or osteophytosis and sclerosis.
Overall, the examiners found that the CBCT and
helical CT were both highly reliable for the evaluation of
the mandibular condyle.
However, side-by-side, the CBCT
images were consistently of superior quality.
None-the-
less, the experienced radiologists were able to accurately
diagnose abnormalities using either system.
The authors
conclude that CBCT being much cheaper and with considerably
lower radiation dose in patients examinations, it is both a
cost effective and a dose effective way to examine the bony
components of the TMJ.
With respect to the literature reviewed, one could
conclude that CBCT is at least as if not more reliable than
helical CT in its imaging capabilities within the oral
maxillofacial complex.
Radiation Dosimetry
One of the greatest of advantages of the CBCT
scanner is its ability to produce high resolution
volumetric images with only a fraction of the radiation
required with helical CT scanners.
This is accomplished
through the use of a cone shaped beam large enough to
25
capture the region of interest in one rotation.
The beam
uses the x-ray emissions very efficiently,8 with less
scatter radiation than conventional CT,7 thereby reducing
the effective dose to the patient.8
Hashimoto et al. found
the exposure dose for the Asterion MDCT to be 400 times
greater than that for the 3DX.
The measured skin dose per
examination was 1.19 mSv for 3DX and 458 mSv for MDCT.6
Mah, Danforth, Bumann, and Hatcher reported that the total
radiation absorbed for a CBCT scan is approximately 20%
that of a conventional CT scan.5
Ludlow, Davies-Ludlow, Brooks, and Howerton
evaluated the dosimetry of 3 CBCT devices: CB Mercuray,
NewTom 3G, and i-CAT.9
These devices were selected for
their capacity to perform 12 inch full field of view (FOV)
examinations.
The 12 inch FOV permits imaging of the full
anatomic region used in craniometric calculations for
orthodontic diagnosis and treatment planning.
Termoluminescent dosemeters (TLDs) were placed in 24 sites
throughout the layers of a tissue-equivalent human skull
RANDO phantom.
The locations of the sensors reflected
critical organs known to be sensitive to radiation.
Ludlow et al.9 found that the dose varied
substantially depending on the device.
Effective doses
were derived from calculations in accordance with both
26
International Commission on Radiological Protection (ICRP)
1990 tissue weights and proposed 2005 tissue weights.
The
effective doses in mSv were: NewTom 3G (45, 59), i-CAT
(135, 193), and CB Mercuray (477, 558).
For the i-CAT
examination doses were 3 to 3.3 times greater than the
NewTom 3G; and the CB Mercuray was 10.7 to 9.5 greater than
the NewTom 3G.
The FOV and various other factors play a
role in determining the radiation dosimetry.
Some of these
include exposure settings such as kV, mA, and exposure
time.
For the i-CAT these settings are fixed.
The NewTom
3G, on the other hand, is capable of adjusting its mA
according to the size of the patient.
The machine is able
to determine proper settings by taking a scout scan prior
to the main exposure.
Lastly, the Mercuray is adjustable
at the discretion of the operator.
The three systems
produced effective doses 4 to 42 times greater than
comparable panoramic examination doses (6.3 mSv, 13.3 mSv).9
Though the full FOV doses from the CBCT units used in this
study were 2-23% of comparable conventional CT examinations
reported in the literature,9 it is still important for the
clinician to recognize the fact that the radiation doses
are much greater than that of conventional radiography.
There is no question that CBCT imaging provides far
more undistorted, accurate information than conventional
27
radiography.
However, unless the clinician is taking full
advantage of the added diagnostic capabilities in CBCT
examinations, they should consider using techniques that
expose the patient to less radiation in accordance with the
ALARA principle (As Low As Reasonably Achievable).
Farman
and Scarfe41 suggest an imaging method which utilizes the
CBCT scout radiograph as 2D lateral cephalogram for
cephalometric analysis; and then using this information in
determining more specific regions requiring full 3D
renditions.
A more focused FOV would minimize radiation
exposure to the patient.
Chidiac, Shofer, Al-Kutoubi,
Laster and Ghafari42 also examined the possibility of
conventional CT scout radiographs being used for
cephalometric analysis. They found that CT scouts and
conventional cephalograms were close in angular
measurements, but differed in accuracy of imaging linear
measurements.
They concluded that logistic and economic
considerations favor the use of conventional cephalograms.
It is important to take into account that these
considerations may not be as applicable with CBCT.
As CBCT
technology progresses, exposure time and radiation dosages
are becoming less, and image quality is increasing.
28
Size and Cost
CBCT scanners are substantially lighter and smaller
than conventional CT scanners.
They have no special
electrical requirements, no floor strengthening is
required, a there is no need for room cooling during
operation.38
Unlike CBCT, the conventional CT scanners do
not lend themselves to miniaturization because of the space
required for the fan shaped beam to spiral around the
entire body.3
The CBCT scanners are built specifically for
oral and maxillofacial imaging; and the FOV of the cone
beam allows the entire complex to be scanned in one
revolution.
Thereby, allowing the CBCT to be more compact,
and therefore resulting in a smaller footprint requirement.
Because the head and neck can be adequately stabilized,
speed of the beam revolution is not as critical for CBCT as
it is for conventional CT.
The conventional CT fan beam
must move at very high speeds in order to avoid artifacts
produced by movement of the heart, lungs, and bowels.3
The
mechanisms required to move the beam at high speeds while
at the same time moving the patient through the scanner,
can add significant cost to the design of the conventional
CT machine.3
Another factor that may cause a price
differential is that the use of a cone beam over a fan
beam, significantly increases x-ray utilization, thereby
29
reducing the x-ray tube heat capacity required for
volumetric scanning.3
CBCT Orthodontic Applications
Since the introduction of CBCT in the late 1990s,
it has become well established as an effective radiographic
tool for oral and maxillofacial diagnosis.
CBCT is being
utilized for many of the same applications CT has been used
for in the past.
However, with its improved
characteristics, such as lower radiation and improved
accessibility and affordability, it is being employed to a
much greater degree in orthodontic diagnosis and treatment
planning.
Accuracy
Cephalometric radiography has been the standard for
the assessment of skeletal, dental, and soft tissue
relationships since its development in the early 1900s.
Cephalometrics is used to describe craniofacial morphology,
evaluate growth, treatment plan, and evaluate treatment
results.
One of the great shortcomings of the lateral
cephalogram is that it is a 2D representation of a 3D
structure.
Two-dimensional images are not a good
30
representation of the patient’s 3D anatomic truth.
No
individual’s face is completely symmetric; and the lack of
accurate superimposition of these asymmetric halves creates
error in landmark identification43 and skews cephalometric
measurements.
Measurement error also results from the
magnification produced in conventional cephalometry.
CT
and CBCT technology makes it possible to create
anatomically true (1:1 in size) images devoid of
magnification and superimposition.
Apart from the error
that results from operator landmark identification,43 this
significantly reduces the error in linear and geometric
measurements.11,12
Enlow, in 2000, says this about the
future of cephalometric imaging:
“The near-future will be
based on the actual biology of an individual’s own
craniofacial growth and development, and it will be
determined by a three-dimensional evaluation based on that
person’s actual morphogenic characteristics, not simply
developmentally irrelevant radiographic landmarks.”44
Both
the CT and CBCT are able to create accurate 3D
representations of the craniofacial complex, however; CT
has had little representation in orthodontic diagnosis and
treatment planning due to: high cost, elevated radiation,
and difficulty in image interpretation.
31
The accuracy of linear, geometric, and volumetric
measurements has been addressed by a number of
investigators.
A 2004 study by Lascala, Panella, and
Marques evaluated the accuracy of the linear measurements
obtained in CBCT images with the NewTom 9000.11
Thirteen
internal and external measurements were made on eight dry
skulls with a digital caliper.
These same measurements
were repeated in CBCT examinations and compared with the
real measurements.
They found that all CBCT measurements
were slightly underestimated relative to the real
measurements, but only statistically significant at the
base of the skull.
A possible explanation for the
measurement variability is that most of the measurements
were taken outside of the dentomaxillary area, which is the
region CBCT scanners are designed to image.
Therefore, the
authors concluded that CBCT is reliable for linear
evaluation measurements of structures closely associated
with dental and maxillofacial imaging.11
Kobayashi, Shimoda, Nakagawa, and Yamamoto,45 also
investigated the accuracy of linear measurements with CBCT,
focusing on dentomaxillary structures alone.
Both CBCT and
CT measurements were compared with actual digital caliper
measurements made on sliced cadaver mandibles.
They found
that CBCT could be used to measure the distance between two
32
points in the mandible more accurately than with CT.
The
measurement error was found to range from 0.01 to
0.65 mm on the CBCT and 0 to 1.11 mm for images produced by
the CT.
The increased measurement error in the CT images
may be due to the loss of resolution of the system in the
direction of reformatting.45
Overall, it was concluded
that, CBCT proved to be a reliable tool for preoperative
evaluation before dental implant surgery due to its high
resolution, low cost, and low radiation.
It is also important to study the accuracy of CBCT
from a geometric point of view.
Marmulla, Wörtche,
Mühling, and Hassfeld12 did just that in their 2005 article:
Geometric accuracy of the NewTom 9000 Cone Beam CT.
They
found that the NewTom CBCT scanner could produce volume
tomograms whose geometric distortion was below the
resolution power of the volume tomograph.
A maximum
deviation of 0.3 mm was determined from the 216
measurements performed on a polymethylmethacrylate block of
known dimensions.
The measurement accuracy of CBCT has been
investigated on an even smaller scale through the
evaluation of periodontal defects.
Misch, Yi, and Sarment46
found that CBCT measurement were as equally useful for the
evaluation of interproximal defects as traditional methods:
33
periodontal probing and periapical radiographs. However,
CBCT has a distinct advantage over conventional periapical
radiographs in that it allows the clinician to accurately
evaluate the buccal and lingual bony defects as well.
With respect to the current study, it is important
to consider the capabilities of the CBCT to accurately
represent the TMJ in volumetric reconstructions.
A recent
study by Hilgers, Scarfe, Scheetz, and Farman4 set out to
investigate the accuracy of CBCT TMJ measurements. The
purpose of the study was to develop CBCT MPR projections
that depict TMJ morphology and select mandibular
relationships, and then compare the reliability and
accuracy of CBCT measurements with conventional
cephalograms in three planes (lateral, posteroanterior, and
submentovertex).
TMJ articulations of twenty-five dry
skulls were inspected using digital calipers for direct
measurements, and the imaging modalities mentioned above
for radiographic examination.
They found that the i-CAT
CBCT accurately depicted the TMJ in all dimensions; and was
significantly more accurate than the conventional
cephalograms in all three orthogonal planes.
34
Impacted Teeth and Oral Abnormalities
Ectopic cuspids are a relatively common occurrence
the orthodontist must address.
The ectopic or impacted
incidences of maxillary cuspids is only second to that of
the third molar; and has been reported by various authors
in ranges that fall between 0.92% and 3%.10,47
Impactions
are twice as common in females as in males.47
Maxillary
impactions are most often located palatally (85%);10 and of
the patients with maxillary impactions, approximately 8%
are bilateral.47
The prevalence of mandibular impactions
(0.35%)47 is much lower than that of maxillary impactions.
The first challenge in treating impacted or ectopic
cuspids is determining their location.
The most common way
of identifying buccal-lingual position of an object has
been through the use of the buccal object rule; also known
as the tube shift technique or Clark’s rule.
This is
accomplished by taking two or more radiographs at different
projection angles.
The objects position is then identified
by the relative positions of two separate objects changing
as the projection angle changes.
This method has proven to
be 92% accurate in identifying the location of impacted or
ectopically erupting maxillary cuspids.48
Another method is
to use two radiographs taken at right angles to each other
such as a periapical and occlusal film.
35
Not only do ectopically erupting cuspids lead to
impaction, but they can also lead to resorption of the
neighboring permanent teeth.
In 1987 Ericson and Kurol48
reported a 0.7% prevalence of resorbed permanent incisors
due to ectopic eruption of maxillary cuspids.
Most
reported numbers of resorptions are relatively low,
however, it has been suggested that they occur more often
than is generally assumed.49
This may be due to the method
of radiographic analysis used in these studies.
Often, 2D
images are incapable of revealing adequate detail needed to
make these diagnoses.
Much of this is due to the overlap
of the incisors by the ectopic canine.50
With the use of 3D imaging, Ericson and Kurol, in
2000,50 discovered that 93% of ectopically positioned
canines were in direct contact with the root of the
adjacent lateral incisor and 12% with the central incisor.
CT scanning substantially increased the detection of
incisor root resorptions.
In fact, 48% of the subjects in
their study had resorption of maxillary incisors due to the
ectopic eruption of maxillary canines.
In this same study
intraoral films and CT scans were compared for their
diagnostic abilities in revealing maxillary incisor
resorptions.
The number of root resorptions on lateral
incisors increased by 53% with the use of the CT imaging
36
over intraoral radiographs.50
With this in mind, it becomes
important to consider the diagnostic advantages CT scanning
can afford the clinician in management of ectopic canines.
Though this technology has been available for some time, it
is seldom used due to issues related to cost, risk/benefit,
access, and expertise in reading the CT.10
The introduction of CBCT has made 3D imaging more
amenable for the evaluation of ectopic erupting and/or
impacted canines.
A more recent study by Walker, Enciso,
and Mah10 supports the claims of incisor resorption
prevalence made in the previous study.50
For this study
CBCT was used instead of conventional CT; and it proved to
be equally effective in: locating maxillary ectopic
cuspids, defining their proximity to adjacent teeth, and
identifying and quantifying the extent of root resorption
caused by ectopic erupting canines.
CBCT can also be quite useful in the detection of
other oral abnormalities.
Some clinicians across the USA
have begun to use CBCT in routine dental examination
procedures.
Initial reports from these clinicians have
revealed a higher incidence of oral abnormalities than
previously suspected (i.e. oral cysts, ectopic/buried teeth
and supernumeraries).7
Kau et al. recommend that:
37
The value of these findings must be taken with
caution, as the number of elective treatments that
may be carried out may be limited. This leads to the
question of whether to intervene in every abnormality
located on these three-dimensional images and the
extent to which the patient needs to be informed. In
the event that these abnormalities were to lead to
pathological episodes, what responsibilities would
the clinician hold in the decision making process?
This could lead to a host of future medico-legal
problems on how clinicians and patients manage
information.7
With the increasing availability and ever improving
technological capabilities, CBCT is becoming more and more
prevalent in the orthodontic office.
With respect to the
above statement, the practicing clinician will have to make
the decision as to when it is appropriate to utilize this
technology; and when abnormalities are revealed in so
doing, what actions will they take, or will they be
expected to take in their management?
Airway Analysis
Evaluation of the upper airway can be accomplished
with a variety of different imaging modalities.
Historically, the most conventional means for the
orthodontist has been through the use of lateral
cephalograms.
However, its diagnostic capabilities are
limited due to its inability to accurately represent the 3D
anatomy of the skeletal and soft tissue structures that
38
comprise the oral and nasopharyngeal airways.
Proper
functionality of the airway relies on adequate volumes of
air to be allowed to pass through (airflow).
One of the
main goals in airway evaluation is to define the volumetric
capabilities of the airway under observation.
Lateral
cephalometric radiographs have no volumetric capabilities
to assist in this task.
Conventional CT scans do have this
capacity; however, they typically demonstrate poor
resolution for upper airway adipose tissue when compared to
magnetic resonance imaging (MRI).
CBCT volumetric imaging abilities may also be
useful in the evaluation of the airway.
With this being a
newer technology, few studies have been accomplished to
examine its capabilities in this aspect.
Aboudara,
Hatcher, Nielsen, and Miller51 conducted a retrospective
cross-sectional chart review pilot study that evaluated the
2D airway from lateral cephalograms and 3D airway structure
from CBCT scans.
They found that there was intra-subject
variation in the airway CT volume to lateral Cephalogram
area; and the variation was greater in airway volume than
airway area.
They concluded that there may be airway
information that is not accurately depicted with the
lateral cephalogram and that more analysis with a greater
sample size is needed. In a more recent study CBCT was used
39
to evaluate volumetric changes that occur in the airway
following mandibular advancement or setback surgery.52
Contrary to what earlier studies using lateral cephalograms
have shown, CBCT examination did not reveal a significant
change in airway volume with advancement or setback
procedures.52
This difference in findings is likely due to
the inability of the lateral cephalogram to measure the
changes that occur in all directions as is possible with
CBCT.52
Orthognathic Surgery
Many of the applications of CBCT in conventional
orthodontics also apply in combined orthodonticorthognathic surgery treatments.
In fact, 3D CT has
already been applied to a much greater extent in
maxillofacial surgery than in orthodontics.
Conventional
CT has been widely used in surgical planning for
craniofacial malformations, acquired defects, skull base
abnormalities, head and neck cancer; it has also found its
place in the measurement of the volume of oral tumors,
analysis of primary nasal deformity in cleft lip and palate
infants, assessment of naso-orbitoethmoidal fractures, and
for the evaluation of airway changes as well as for in
vitro experimental validation of 3D landmark measurement in
40
craniofacial surgery planning.32
Conventional CT and CBCT
provide the surgeon with the ability to create Biomodels
through stereolithography or milling process.
Biomodels
provide the surgeon with invaluable information to assist
in presurgical planning for orthognathic cases, traumatic
injury cases, as well as a variety of other applications.
Temporomandibular Joint Evaluation: Imaging
The summer of 1987 a Michigan jury awarded $850,000
in an orthodontic case.53
The plaintiff, a teenage female,
received routine orthodontic treatment to correct a Class
II Division 1 malocclusion with a 7 mm anterior open bite.
The patient was treated with two upper bicuspid extractions
and the use of headgear.
The patient experienced no signs
or symptoms of temporomandibular disorder (TMD) prior to
treatment, but developed symptoms shortly after completion
of treatment.
The expert witnesses for the plaintiff
agreed that the patient’s predilection for TMD should have
been identified prior to initiating orthodontic care, that
the extraction of bicuspids was contraindicated and was a
departure from the standard of acceptable orthodontic
practice.
41
Publicity of the court decision sparked great
interest in the relationship between orthodontic treatment
and TMD in the dental research community.
Studies
conducted since and have demonstrated limited evidence of a
relationship between TMD and orthodontic treatment.54-56
Kim, Graber and Viana54 performed a meta-analysis of the
literature, using the evidence from 31 primary studies to
analyze and evaluate the relationship between orthodontic
treatment and TMD.
The data from their meta-analysis
showed no indication that traditional orthodontic treatment
increased the prevalence of TMD.
In another review of the
literature, Luther55 found that orthodontic treatment has
little role to play in worsening or precipitating TMD when
treated patients are compared with untreated individuals.
Studies suggest that orthodontic treatment is at worst, TMD
neutral.55
Despite the lack of evidence to establish a link
between orthodontic treatment and the incidence of TMD;
orthodontists must continue to be vigilant in their
pretreatment evaluations, make accurate TMJ diagnoses, and
be quick to respond when unanticipated joint problems
arise.
A proper screening for TMJ health status requires a
thorough history, as well as comprehensive clinical and
42
radiographic examinations.
TMJ radiographic examination
will be the focus for the purposes of the current study.
According to Okeson57 there are three limiting
conditions that must be considered prior interpretation of
TMJ radiographs: 1) Absence of articular surfaces, 2)
Superimposition of subarticular surfaces, and 3) Variations
in normal.
1) Absence of articular surfaces:
The
articular surfaces of the condyle, disk, and fossa are made
of dense fibrous connective tissue standard radiographs
have no capacity to capture.
Therefore, the surfaces of
the condyle and fossa viewed in standard radiographs are
actually subarticular bone.
subarticular surfaces:
2) Superimpositions of
Apart from corrected tomograms and
CT scans, routine radiographic images of the TMJ preset
with superimpositions of subarticular surfaces, limiting
the application of these images. 3)
Variations in normal:
Variation from what one considers normal TMJ morphology is
not always an indication of pathosis.
The examiner must
take into account the great variation that exists from one
subject to another.
One must also consider other factors
that may contribute such as radiographic technique and
patient positioning.
Okeson57 recommends that radiographs
should not be used to diagnose TMJ disorder, but instead,
43
serve as additional information to support or negate an
already established clinical diagnosis.
2D Imaging
Various forms of imaging techniques can be used to
evaluate temporomandibular joint (TMJ) form and function.
Okeson57 suggests four basic radiographic techniques that
can be used in most general dental practices: the
panoramic, lateral transcranial, transpharyneal, and
transmaxillary anteroposterior (AP) views.
Generally, in
orthodontic diagnosis and treatment planning, the panoramic
radiograph serves as an initial screening tool for TMJ
evaluation.
Dixon30 suggests that in consideration of other
imaging techniques for detecting osseous abnormalities, the
panoramic radiograph is an excellent choice for a screening
view of the TMJ due to its cost effectiveness, high
availability, and relatively low radiation dose.
However,
the diagnostic capability of panoramic radiographs is
limited to gross osseous changes.1
Only obvious erosions,
sclerosis, and osteophytes of the condyle can be
identified.1
It has limited use for the identification of
early lesions, and no capability to provide information on
joint soft tissue status.30
44
Conventional tomography may give the practitioner
more insight into the exact morphology of the osseous
structures of the TMJ.
Generally it is accepted as being
superior to plane film radiography in the assessment of
joint spaces and the detection of osseous lesions of the
joint complex.30
Also, tomography has been shown to reveal
a greater number of structural changes, and represent
anatomic structures more accurately than transcranial
radiography.58-60
However, like the panoramic radiograph, it
has limited utility in the detection of early arthritic
changes; and even more advanced changes in the fossa tend
to go undetected.30
Dixon30 suggests that “this may explain
why radiographic findings often correlate poorly with
clinical signs and symptoms, which often peak early in the
disease”.
Brookes et al.1 makes these final remarks about
conventional tomography:
Because conventional tomography has been used so
often for years, it is relatively well studied. It
is clear that there are few if any correlations
between clinical and radiographic findings. Thus
patients with radiographically normal joints may have
pain, just as those with clear signs of bone
abnormalities may be pain free. The value of
radiographic evaluation lies in the ability of
radiographs to help clinicians better understand the
condition of the joint. Indeed, some studies have
indicated that tomography can provide unanticipated
information that may lead to a change in treatment
plan. In contrast, however, other investigations
have concluded that the presence or extent of
radiographic signs of osseous pathoses are of little
45
prognostic value in the outcome of treatment and that
tomography has little effect on the diagnosis or
treatment plan of patients with TMJ disorders.
Similar to plane film radiography, conventional
tomography has little or no capacity to image soft tissue
structures within the joint space; nor can it reliably
predict articular disk position.1
Arthrography and
arthrotomography, contrast-enhanced plain or tomographic
radiography, augment the soft tissue imaging potential of
plane and tomographic radiography by introducing a
radiopaque contrast medium into the lower, upper, or both
joint space compartments.
Often, the radiopaque medium is
injected under fluoroscopic guidance.
Once contrast medium
injection is accomplished, plane films or tomograms can
then be obtained in the usual manner.
The articular disk
appears as a radiolucent mass against a background of
contrast medium; thereby, providing information on disk
location, shape, and even movement with the use of video
fluoroscopy.
Disk perforations, attachments, and or disk
adhesions and capsular tears may also be determined by flow
of contrast medium from one joint space compartment to
another.1
Arthrography is an excellent, mildly invasive
modality for radiographic evaluation of joint soft tissue
46
dynamics and internal joint derangements.30
However, there
are a number of disadvantages to Arthrography which include
but are not limited to:
patient discomfort associated with
injection of contrast medium;30 true joint dynamics may be
disturbed by the presence of contrast medium;30 extensive
experience is required for such a technique sensitive
procedure;30 possible allergic reactions to the contrast
medium;1 and radiation exposure to the patient can be
significant, particularly with the use of fluoroscopy.1
Arthrography itself yields little information about the
osseous structures of the TMJ due to the presence of
radiopaque contrast medium.1
Therefore, another form of
imaging must be used in conjunction with arthrography in
order to evaluate the osseous and soft tissue structures.
3D Imaging
With CT and CBCT it is possible to image both hard
and soft tissues without the introduction of contrast
medium.
Thereby, allowing the practitioner to assess the
disk-condyle relationship without disturbing the existing
anatomic relationship, or causing physical trauma to the
local tissues.57
Honda et al.61 recommend that air contrast
arthrography using CBCT is effective for diagnosis of disk
location, configuration, presence of adhesions and
47
perforations; and should be considered for cases of TMJ
disorder in the presence of contrast medium
contraindication or MRI contraindications.
“Though some
early studies showed promise for CT in the detection of
internal derangement, magnetic resonance imaging (MRI) has
proven to demonstrate disk position and morphologic
condition better than CT and has almost completely
supplanted CT for this diagnostic purpose.”1
Today, MRI is
the preferred examination for TMJ soft tissue pathology.61
Dixon30 reports that several studies have assessed the
validity of MRI in diagnosing disk displacements and found
that the sensitivity in all studies was at least 0.86.
One
study reported an accuracy that may reach as high as 95%.62
MRI has also been shown to be useful in the evaluation of
osseous lesions, and joint effusion; and shows potential as
a diagnostic aid for avascular necrosis of the condyle.30
One of the chief advantages of MRI over
radiographic techniques lies in the substitution of
superconducting magnets and radio wave energy for the well
know hazards of ionizing radiation.30
safe for everyone.
However, MRI is not
The powerful magnetic fields produced
in the imaging process can have highly detrimental effects
on patients that have metal artifacts within their body,
48
such as:
pace makers, intracranial vascular clips, and
metal particles in the eye or other vital structures.1
CT and CBCT are best suited for the evaluation of
osseous morphology.1
They serve well for the diagnosis of
bony abnormalities including fractures, dislocations,
arthritis, ankylosis, and neoplasia.1
They can also provide
accurate information on the position of the condyle within
the fossa in open and closed mouth conditions.63
The 1997
position paper of the American Academy of Oral and
Maxillofacial Radiology1 suggests that, because of its high
cost and radiation dose, CT of the TMJ be reserved for the
evaluation of foreign body giant cell reaction to silicon
or polytetrafluoroethylene sheet implants, suspected
tumors, ankylosis, and complex facial fractures.
However,
since the introduction of cone-beam technology in 2000,
these contraindications have been greatly diminished. For
this reason one might consider using CBCT to evaluate
conditions of lesser severity.
Some orthodontists have
already begun to use CBCT as a comprehensive radiographic
examination for diagnosis and treatment planning.
In a
feature article from the American Association of Dental
Maxillofacial Radiographic Technicians, Way64 shares his
CBCT imaging protocols and viewing sequences for the
orthodontist.
He also describes patient flow and
49
integration of CBCT into the orthodontic initial
examination process.
CBCT viewing involves screening for
pathology, airway analysis, ectopically erupting as well as
impacted teeth, asymmetries, cephalometric analysis and TMJ
studies.64
A 2004 report by Tsiklakis, Syriopoulos and
Stamatakis63 describes a reconstruction technique for
radiographic examination of the TMJ using CBCT.
The
technique results in obtaining lateral and coronal CBCT
images as well as 3D reconstructions of the TMJ.
To assess
range and type of condylar movement a second scan was made
with the patient’s mouth open.
This procedure was employed
for four case studies presented in the report.
Tsiklakis
et al. concluded that the technique provided a complete
radiographic investigation of the bony components of the
TMJ, reconstructed images were of high diagnostic quality,
the scanning time and radiation dose were smaller than that
of conventional CT, and therefore, should be considered the
imaging modality of choice for the examination of bony
changes in the TMJ.
50
Hounsfield Units as a Measure of Bone Density
Computed Tomography produces raw data by measuring
the extent of the x-ray transmission through various
tissues.
This data is interpreted by computer software and
for each given pixel a relative attenuation coefficient is
determined.
The attenuation coefficients are normalized
with respect to water as the reference material, and a
magnifying constant is then applied to produce a CT
number.18
These CT numbers are often described in terms of
Hounsfield units (HU).
Hounsfield19 originally described
the HU as an absorption value; and he constructed a scale
(see Figure 2.5) to demonstrate the accuracy to which the
absorption values could be ascertained on a visual image.
For the machine he described,19 the scale ranged from air
(-1000) at the bottom of the scale, to bone (1000) at the
top of the scale.
For convenience, water was chosen to be
0 at the center of the scale.
This was done because the
absorption of water is close to that of soft tissue.
The
range of CT numbers is 2000 HU wide; yet some modern
scanners have a greater range up to 4000 HU.
Each number
represents a different shade of gray within the spectrum.
The range of tones between black and white seen in an image
can be limited to a large or small window within the scale.
51
Figure 2.5. Scale used by Hounsfield to demonstrate the
accuracy to which absorption values can be ascertained on
the CT picture.19
This window can then be raised or lowered depending upon
the absorption value of the material of interest.19
The
central HU of all the numbers within the window is referred
to as the window level.
The window width displays the
tissues of interest in various shades of gray.
All tissues
outside this range are displayed as either black or white.31
Figure 2.6 displays window widths for various tissues in
the body.
It suggests that the ideal window width for
viewing bone would fall between +400 and +1000 HU.
52
Figure 2.6. Hounsfield Scale of CT Numbers (image taken
from Jackson).31
The Hounsfield Unit has been used to describe physical
density and achieve reasonable volume estimates of anatomic
structures.13-15,18,65,66
The greater a tissue density, the
higher the HU numbers will be within the respective window
width.
In a CT study aimed at characterization of adrenal
masses, Nawariaku et al.67 found that one could
differentiate adrenal adenomas from non-adenomas by using
an attenuation value of ≤10 HU on non-enhanced conventional
CT.
A number of studies have been performed in efforts to
quantify bone density based on Hounsfield values.13-15
of these studies were accomplished in order to help
classify bone types best suitable to support dental
implants.13,14,17
Pre-implant evaluation of bone density
helps the surgeon to identify suitable implant sites,
thereby improving surgical planning, and potentially
53
Many
improving implant success rates.13
Aranyarachkul et al.13
examined variations in bone density in designated implant
recipient sites using both CBCT and conventional CT.
They
found both modalities to be consistent in their
measurements of bone density value; however, the values
were generally higher for CBCT.
Whether CBCT or
conventional CT values are closer to corresponding
histological bone densities has yet to be investigated.
Yet, other studies have focused on relating CT
numbers not only to the density, but also the mechanical
properties of bone.15,16
Stoppie et al.15 and Rho et al.16
found that CT predictions of the mechanical properties in
bone were most accurate for specimens of full trabecular
bone, with little or no cortical bone present.
For these,
a good correlation was found between the HU value and bone
mineral density (BMD).
For jaws with a thicker cortical
layer the HU values were less reliable as predictors of
mechanical properties.
Bone tissue density is a function
of both porosity and mineralization.16
Rho et al.16 suggest
that because such a poor correlation exists between CT
numbers and cortical bone density; and a much higher
correlation between CT numbers and cancellous bone density
is evident, it might be reasonable to consider CT numbers
54
to be more of a function of bone porosity than
mineralization.
CT numbers cannot be accepted as an absolute for
characterization of a tissue type or lesion.68
CT numbers
may vary significantly from one scanner to another, or even
between two scanners of the same make and model.68
Consequently, one may not assume that CT numbers reported
from one scanner will transfer over to another for tissue
identification.68
Stadler et al.69 found significant
differences in CT density measurements in adrenal tumors
with the use of three different CT scanners, and or imaging
protocols.
With some degree of success, standardized
calibration methods have been employed in efforts to
minimize inter-scan discrepancies.70
Cann71 evaluated
sixteen conventional CT scanners from eight different
manufacturers for reproducibility of CT numbers.
It was
concluded that not all CT scanners are equally suitable for
quantitative CT applications, and the results of one
scanner may not be comparable to those of another unless
correction factors are applied.71
With attention to detail,
strict standardization in all parameters, continual
manufacturer support, and application of proper calibration
methods, reproducibility can be optimized.71
55
CT has also been utilized in defining volumes of
various tissues within a subject.66
Breiman et al.66 found
in vivo volume estimations obtained with CT to be
comparable to those determined by water displacement.
The
authors suggest that CT potentially offers the most
accurate modality for the estimation of in vivo volumes.
This has also been applied in efforts to identify
osteoporotic areas within bone.
CT imaging offers a myriad of potential diagnostic
applications through its density and volumetric assessment
capabilities; however, its best application for diagnosing
bone pathosis, and or anatomical variation, remains to be
through the visual examination of CT reconstructions.
These reconstructions can be displayed as three-dimensional
volumetric renderings, revealing all external surfaces of
the tissue of interest; or as cross-sectional images which
represent a limited area of the internal and external
anatomy.
The former of the two offers a potentially more
simplistic and efficient manner in which to examine the
external surfaces of the condyle.
Within the condyle there is variation in bone
density and composition. Cortical bone, trabeculae, and
intertrabecular tissue have varying densities and
mechanical properties.
These differences present a
56
challenge when examining the bony surfaces of the condyle
with the CT 3D reconstruction.
The examiner must decide
what window level and width will most accurately represent
the anatomical truth.
It is important for the clinician to
be able to reliably detect osseous abnormalities such as
cortical erosion, osteophytosis, osteoarthritis, and
sclerosis.
The following images demonstrate how varying
window widths and levels can profoundly affect the
appearance of CBCT reconstructions.
Within Figures 2.7 to
2.11 there are 5 sets of images with different window
widths and levels ranging from 197 HU to -600 HU.
Each set
includes the all of the bone densities from the above
mentioned numbers up to 2476 HU.
V-works™4.0 imaging
software recommends that the window width and level for
viewing osseous structures is between 176 HU and 2476 HU72
(Table 2.1).
The images of Figure 2.7 were reconstructed
using a window width of 2279 HU and window level at 1140
HU.
This gives a range from 197 HU to 2476 HU; one very
similar to that recommended by V-works™4.0.
Everything
outside this range will be represented by black or white.
White represents high density and black low density.
Density values of a material are not directly identified
57
Figure 2.7.
3D reconstruction (197 HU to 2476 HU)
Figure 2.8.
3D reconstruction (-100 HU to 2476 HU)
Figure 2.9.
3D reconstruction (-200 HU to 2476 HU)
58
Figure 2.10.
3D reconstruction (-400 HU to 2476 HU)
Figure 2.11.
3D reconstruction (-600 HU to 2476 HU)
with Hounsfield units.
For the V-works™ 4.0 imaging
software the density values start with the low density
value of air being at 0 and peak with a high density value
of 4095 for metal. In order to convert density values to
HU, one must simply subtract 1024 from the density value.72
Figure 2.12 reveals the photographic appearance of the
right condyle shown in the CBCT reconstructions of
59
Figures 2.7 - 2.11.
When compared with Figure 2.7, there
is a substantial difference in the condylar anatomy.
While
on the other hand, the majority of the mandibular bony
anatomy seems to be well represented in Figure 2.7.
As the
window widths widen and extend below the bone range, as in
figures 2.7-2.11, the CBCT images of the condyle approach
a)
b)
c)
d)
Figure 2.12. These photographs of the mandible shown in
the previous reconstruction images demonstrate the true
appearance of the articular surfaces of the right condyle.
a) lateral pole, b) posterior surface, c) right mandible,
d) posterior/lingual view right half of the mandible
60
Table 2.1. Window widths for density values and Hounsfield
numbers in the body.72
a visual level closer to that of the photographic images.
Of the five reconstructions the -600 to 2476 gives the most
accurate representation of the bony articular anatomy.
It
is likely that there is a greater range of densities within
the condyle when compared to the rest of the mandible, and
that it is heavily weighted on the low density end of the
scale.
This presents a problem when examining the condyle
in vivo.
These density ranges extend into the soft tissue
range; and the presence soft tissues begins to occlude
one’s capacity to view the osseous anatomy with the 3D
reconstruction.
The purpose of the present study is to
investigate what the ideal Hounsfield unit values are for
the examination of the condyle; and if an ideal window
width and level can be identified, can one reliably
evaluate the mandibular condyle using the CBCT 3D
reconstruction?
61
References
1. Brooks SL, Brand JW, Gibbs SJ, Hollender L, Lurie AG,
Omnell KA et al. Imaging of the temporomandibular joint: a
position paper of the American Academy of Oral and
Maxillofacial Radiology. Oral Surg Oral Med Oral Pathol
Oral Radiol Endod 1997;83:609-618.
2. Quintero JC, Trosien A, Hatcher D, Kapila S.
Craniofacial imaging in orthodontics: historical
perspective, current status, and future developments. Angle
Orthod 1999;69:491-506.
3. Sukovic P. Cone beam computed tomography in craniofacial
imaging. Orthod Craniofac Res 2003;6 Suppl 1:31-36;
discussion 179-182.
4. Hilgers ML, Scarfe WC, Scheetz JP, Farman AG. Accuracy
of linear temporomandibular joint measurements with cone
beam computed tomography and digital cephalometric
radiography. Am J Orthod Dentofacial Orthop 2005;128:803811.
5. Mah JK, Danforth RA, Bumann A, Hatcher D. Radiation
absorbed in maxillofacial imaging with a new dental
computed tomography device. Oral Surg Oral Med Oral Pathol
Oral Radiol Endod 2003;96:508-513.
6. Honda K, Larheim TA, Maruhashi K, Matsumoto K, Iwai K.
Osseous abnormalities of the mandibular condyle: diagnostic
reliability of cone beam computed tomography compared with
helical computed tomography based on an autopsy material.
Dentomaxillofac Radiol 2006;35:152-157.
7. Kau CH, Richmond S, Palomo JM, Hans MG. Threedimensional cone beam computerized tomography in
orthodontics. J Orthod 2005;32:282-293.
8. Hatcher DC, Aboudara CL. Diagnosis goes digital. Am J
Orthod Dentofacial Orthop 2004;125:512-515.
62
9. Ludlow JB, Davies-Ludlow LE, Brooks SL, Howerton WB.
Dosimetry of 3 CBCT devices for oral and maxillofacial
radiology: CB Mercuray, NewTom 3G and i-CAT.
Dentomaxillofac Radiol 2006;35:219-226.
10. Walker L, Enciso R, Mah J. Three-dimensional
localization of maxillary canines with cone-beam computed
tomography. Am J Orthod Dentofacial Orthop 2005;128:418423.
11. Lascala CA, Panella J, Marques MM. Analysis of the
accuracy of linear measurements obtained by cone beam
computed tomography (CBCT-NewTom). Dentomaxillofac Radiol
2004;33:291-294.
12. Marmulla R, Wortche R, Muhling J, Hassfeld S. Geometric
accuracy of the NewTom 9000 Cone Beam CT. Dentomaxillofac
Radiol 2005;34:28-31.
13. Aranyarachkul P, Caruso J, Gantes B, Schulz E, Riggs M,
Dus I et al. Bone density assessments of dental implant
sites: 2. Quantitative cone-beam computerized tomography.
Int J Oral Maxillofac Implants 2005;20:416-424.
14. Shapurian T, Damoulis PD, Reiser GM, Griffin TJ, Rand
WM. Quantitative evaluation of bone density using the
Hounsfield index. Int J Oral Maxillofac Implants
2006;21:290-297.
15. Stoppie N, Pattijn V, Van Cleynenbreugel T, Wevers M,
Vander Sloten J, Ignace N. Structural and radiological
parameters for the characterization of jawbone. Clin Oral
Implants Res 2006;17:124-133.
16. Rho JY, Hobatho MC, Ashman RB. Relations of mechanical
properties to density and CT numbers in human bone. Med Eng
Phys 1995;17:347-355.
17. Hatcher DC, Dial C, Mayorga C. Cone beam CT for presurgical assessment of implant sites. J Calif Dent Assoc
2003;31:825-833.
63
18. Mull RT. Mass estimates by computed tomography:
physical density from CT numbers. AJR Am J Roentgenol
1984;143:1101-1104.
19. Hounsfield GN. Nobel Award address. Computed medical
imaging. Med Phys 1980;7:283-290.
20. Crawford PR. 100 years of radiology: those early years.
J Can Dent Assoc 1995;61:951-954.
21. Microsoft® Encarta® Online Encyclopedia, X Ray.
http://encarta.msn.com/encyclopedia_761579196/X_Ray.html;
2006. Cited on August 30, 2006.
22. Jacobson PH, Fedran RJ. Making darkness visible: the
discovery of X-ray and its introduction to dentistry. J Am
Dent Assoc 1995;126:1359-1367.
23. Langland OE, Langlais RP. Early pioneers of oral and
maxillofacial radiology. Oral Surg Oral Med Oral Pathol
Oral Radiol Endod 1995;80:496-511.
24. Mupparapu M, Vuppalapati A. Detection of an early
ossification of thyroid cartilage in an adolescent on a
lateral cephalometric radiograph. Angle Orthod 2002;72:576578.
25. Behrents R, Broadbent B, Jr. A Chronological Account of
the Bolton-Brush Growth Studies; 1984.
26. Broadbent B, Jr. A New X-Ray Technique and Its
Application to Orthodontia. The Angle Orthodontist
1931;1:2-24.
27. Hallikainen D. History of panoramic radiography. Acta
Radiol 1996;37:441-445.
28. White S. Pharoah M. Oral Radiology Principles and
Interpretation. St. Louis: Mosby; 2000.
64
29. Hendee WR. Cross sectional medical imaging: a history.
Radiographics 1989;9:1155-1180.
30. Dixon DC. Radiographic diagnosis of temporomandibular
disorders. Semin Orthod 1995;1:207-221.
31. Jackson ST, R. Cross-Sectional Imaging Made Easy.
Edinburgh: Churchill Livingstone; 2004.
32. Papadopoulos MA, Christou PK, Christou PK, Athanasiou
AE, Boettcher P, Zeilhofer HF et al. Three-dimensional
craniofacial reconstruction imaging. Oral Surg Oral Med
Oral Pathol Oral Radiol Endod 2002;93:382-393.
33. Nakajima A, Sameshima GT, Arai Y, Homme Y, Shimizu N,
Dougherty H, Sr. Two- and three-dimensional orthodontic
imaging using limited cone beam-computed tomography. Angle
Orthod 2005;75:895-903.
34. Hashimoto K, Arai Y, Iwai K, Araki M, Kawashima S,
Terakado M. A comparison of a new limited cone beam
computed tomography machine for dental use with a
multidetector row helical CT machine. Oral Surg Oral Med
Oral Pathol Oral Radiol Endod 2003;95:371-377.
35. Araki K, Maki K, Seki K, Sakamaki K, Harata Y, Sakaino
R et al. Characteristics of a newly developed
dentomaxillofacial X-ray cone beam CT scanner (CB
MercuRay): system configuration and physical properties.
Dentomaxillofac Radiol 2004;33:51-59.
36. Hashimoto K, Kawashima S, Araki M, Iwai K, Sawada K,
Akiyama Y. Comparison of image performance between conebeam computed tomography for dental use and four-row
multidetector helical CT. J Oral Sci 2006;48:27-34.
37. Baba R, Ueda K, Okabe M. Using a flat-panel detector in
high resolution cone beam CT for dental imaging.
Dentomaxillofac Radiol 2004;33:285-290.
65
38. 3D Diagnostix inc., Machine comparison chart.
http://www.conebeam.com/cbct-clinician/manufacturers.php;
2006. Cited on March 12, 2006.
39. IMTEC Imaging, Product overview.
http://www.ilumact.com/products.php; 2006. Cited on October
13.
40. Holberg C, Steinhauser S, Geis P, Rudzki-Janson I.
Cone-beam computed tomography in orthodontics: benefits and
limitations. J Orofac Orthop 2005;66:434-444.
41. Farman AG, Scarfe WC. Development of imaging selection
criteria and procedures should precede cephalometric
assessment with cone-beam computed tomography. Am J Orthod
Dentofacial Orthop 2006;130:257-265.
42. Chidiac JJ, Shofer FS, Al-Kutoub A, Laster LL, Ghafari
J. Comparison of CT scanograms and cephalometric
radiographs in craniofacial imaging. Orthod Craniofac Res
2002;5:104-113.
43. Baumrind S, Frantz RC. The reliability of head film
measurements. 1. Landmark identification. Am J Orthod
1971;60:111-127.
44. Enlow D. Discussion. Am J Orthod Dentofacial Orthop
2000;117:147.
45. Kobayashi K, Shimoda S, Nakagawa Y, Yamamoto A.
Accuracy in measurement of distance using limited cone-beam
computerized tomography. Int J Oral Maxillofac Implants
2004;19:228-231.
46. Misch KA, Yi ES, Sarment DP. Accuracy of cone beam
computed tomography for periodontal defect measurements. J
Periodontol 2006;77:1261-1266.
47. Bishara SE. Impacted maxillary canines: a review. Am J
Orthod Dentofacial Orthop 1992;101:159-171.
66
48. Ericson S, Kurol J. Radiographic examination of
ectopically erupting maxillary canines. Am J Orthod
Dentofacial Orthop 1987;91:483-492.
49. Ericson S, Kurol J. Incisor resorption caused by
maxillary cuspids. A radiographic study. Angle Orthod
1987;57:332-346.
50. Ericson S, Kurol PJ. Resorption of incisors after
ectopic eruption of maxillary canines: a CT study. Angle
Orthod 2000;70:415-423.
51. Aboudara CA, Hatcher D, Nielsen IL, Miller A. A threedimensional evaluation of the upper airway in adolescents.
Orthod Craniofac Res 2003;6 Suppl 1:173-175.
52. Stigall M. Oropharyngeal Airway Volume Changes: A
Three-Dimensional Evaluation using Cone-Beam CT in Combined
Orthodontic-Orthognathic Surgery Patients Orthodontics. St.
Louis: Saint Louis University; 2006: p. 76.
53. Pollack B. Cases of note: Michigan jury awards $850,000
in ortho case: a tempest in a teapot. Am J Orthod
Dentofacial Orthop 1988;94:358-360.
54. Kim MR, Graber TM, Viana MA. Orthodontics and
temporomandibular disorder: a meta-analysis. Am J Orthod
Dentofacial Orthop 2002;121:438-446.
55. Luther F. Orthodontics and the temporomandibular joint:
where are we now? Part 1. Orthodontic treatment and
temporomandibular disorders. Angle Orthod 1998;68:295-304.
56. Carlton KL, Nanda RS. Prospective study of
posttreatment changes in the temporomandibular joint. Am J
Orthod Dentofacial Orthop 2002;122:486-490.
57. Okeson J. Management of Temporomandibular Disorders and
Occlusion. St. Louis: Mosby-Year Book, Inc.; 1993.
67
58. Eckerdal O, Lundberg M. The structural situation in
temporomandibular joints. A comparison between conventional
oblique transcranial radiographs, tomograms and histologic
sections. Dentomaxillofac Radiol 1979;8:42-49.
59. Lindvall AM, Helkimo E, Hollender L, Carlsson GE.
Radiographic examination of the temporomandibular joint. A
comparison between radiographic findings and gross and
microscopic morphologic observations. Dentomaxillofac
Radiol 1976;5:24-32.
60. Bean LR, Omnell KA, Oberg T. Comparison between
radiologic observations and macroscopic tissue changes in
temporomandibular joints. Dentomaxillofac Radiol 1977;6:90106.
61. Honda K, Matumoto K, Kashima M, Takano Y, Kawashima S,
Arai Y. Single air contrast arthrography for
temporomandibular joint disorder using limited cone beam
computed tomography for dental use. Dentomaxillofac Radiol
2004;33:271-273.
62. Tasaki MM, Westesson PL. Temporomandibular joint:
diagnostic accuracy with sagittal and coronal MR imaging.
Radiology 1993;186:723-729.
63. Tsiklakis K, Syriopoulos K, Stamatakis HC. Radiographic
examination of the temporomandibular joint using cone beam
computed tomography. Dentomaxillofac Radiol 2004;33:196201.
64. Way D. Utilization of CBCT in an Orthodontic Practice:
AADMRT; 2006.
65. Kobayashi F, Ito J, Hayashi T, Maeda T. A study of
volumetric visualization and quantitative evaluation of
bone trabeculae in helical CT. Dentomaxillofac Radiol
2003;32:181-185.
68
66. Breiman RS, Beck JW, Korobkin M, Glenny R, Akwari OE,
Heaston DK et al. Volume determinations using computed
tomography. AJR Am J Roentgenol 1982;138:329-333.
67. Nwariaku FE, Champine J, Kim LT, Burkey S, O'Keefe G,
Snyder WH, 3rd. Radiologic characterization of adrenal
masses: the role of computed tomography--derived
attenuation values. Surgery 2001;130:1068-1071.
68. Levi C, Gray JE, McCullough EC, Hattery RR. The
unreliability of CT numbers as absolute values. AJR Am J
Roentgenol 1982;139:443-447.
69. Stadler A, Schima W, Prager G, Homolka P, Heinz G,
Saini S et al. CT density measurements for characterization
of adrenal tumors ex vivo: variability among three CT
scanners. AJR Am J Roentgenol 2004;182:671-675.
70. Cann CE. Quantitative CT for determination of bone
mineral density: a review. Radiology 1988;166:509-522.
71. Cann CE. Quantitative CT applications: comparison of
current scanners. Radiology 1987;162:257-261.
72. Cybermed. V-works™4.0. Seoul, Korea: Cybermed, co, Ltd;
2006: p. Imaging software for real time 3D visualization.
69
CHAPTER 3: JOURNAL ARTICLE
Abstract
Cone beam computed tomography (CBCT) has proven to
be a valuable imaging modality for examination of the
temporomandibular joint (TMJ).
CBCT three-dimensional (3D)
reconstructions potentially provide the clinician with
efficient and effective means for evaluation of the
mandibular condyle.
The purpose of the present study is to
investigate what the ideal Hounsfield unit values are for
the examination of the condyle; and if an ideal window
width and level can be identified, can one reliably
evaluate the mandibular condyle using the CBCT 3D
reconstruction?
Linear dimensions between six anatomical sights were
measured with a digital caliper to assess the anatomic
truth for 50 dry human mandibular condyles.
Condyles were
scanned with the i-CAT (Imaging Sciences International,
Hatfield, PA) CBCT; and image reconstruction and assessment
was accomplished using V-works™4.0 (Cybermed Inc, Seoul,
Korea).
Three linear three-dimensional measurements were
made on each of the 50 condyles at eight different
Hounsfield unit (HU) windows.
These measurements were
70
compared with the anatomic truth.
Volumetric measurements
were also completed on all 50 condyles, at 23 different
window levels, in order to define the volumetric
distribution of bone mineral density (BMD) within the
condyle.
Significance testing for 3D linear measurement
and volumetric differences was accomplished using
independent t-tests with a 95% confidence interval.
Significant differences were found in two of the
three linear measurement groups at and below the
recommended viewing window for osseous structures.
The
most accurate measurements were made within the soft tissue
range for HU window levels.
Volumetric distribution
measurements revealed that the condyles were mostly
comprised of low density bone; and that condyles exhibiting
significant changes in linear measurements were shown to
have higher percentages of low density bone than those
condyles with little change from the anatomic truth.
Assessment of the mandibular condyle, using the 3D
reconstruction, is most accurate when accomplished at
density levels below that recommended for osseous
examination.
Utilizing lower window levels, extending into
the soft tissue range, may compromise one’s capacity to
view the bony topography.
This would suggest that CBCT 3D
reconstructed images, by themselves, may not be a reliable
71
means for the diagnosis condylar pathology, and or changes
in condylar morphology.
Introduction
Temporomandibular joint (TMJ) standard radiographic
studies such as the plain film radiography and panoramic
radiography have little capacity to reveal anything more
than gross osseous changes1 within the joint; and therefore,
in some cases a more comprehensive radiographic study is
indicated.
Radiographic analysis of the TMJ is a broad
field and is considered by some to be a separate subset of
oral and maxillofacial radiology,2 consisting of both two
and three-dimensional imaging modalities.
Two-dimensional
(2D) imaging of the TMJ employs conventional radiology to
produce a variety of projections.
The submental-vertex,
lateral transcranial, transpharyneal, transmaxillary (AP),
as well as conventional tomography are a few examples of
the 2D studies used in TMJ evaluation.
Three-dimensional
evaluations, such as computed tomography (CT) and magnetic
resonance imaging (MRI), have been utilized to some degree;
however, historically, high cost,3,4 large radiation
dosage,5,6 large space requirements,3,4 and the high level of
skill required for interpretation have kept its use to a
72
minimum.
With the introduction of limited cone-beam
technology, such deterrents of CT imaging have been greatly
diminished.
With five different cone-beam computed
tomography (CBCT) scanners now available on the world
market, lower radiation dosages,7-10 and lower costs,3 3D
radiography is likely to become more commonplace in the
dental profession.
Demonstrating a broad spectrum of
applications and vastly improved accuracy over 2D
radiography,4,11,12 CBCT proves to be an invaluable diagnostic
tool for the evaluation of the osseous structures of the
TMJ.
Three-dimensional (3D) imaging facilitates
comprehensive examination of the subarticular osseous
surfaces of the TMJ through two different viewing
modalities.
One allows for the tissue of interest to be
examined through progressive crossectional slices in a
chosen plane of orientation; while the other produces 3D
reconstructions capable of being manipulated in any given
direction, thereby facilitating and expediting the
examination procedure.
Often, these two viewing methods
are combined into what is referred to as the multiplanar
reconstruction (MPR).
MPR viewing mode enables the
examiner to simultaneously assess crossectional and threedimensionally reconstructed images (see Figure 3.1).
73
Within the condyle there is variation in bone
density and composition. Cortical bone, trabeculae, and
intertrabecular tissue have varying densities and
mechanical properties.13-17
These differences present a
Figure 3.1. A multiplanar reconstruction (MPR) with axial,
coronal, and sagittal crossectional views, and 3D
reconstruction.
challenge when examining the bony subarticular surfaces of
the condyle with 3D CT imaging.
For computed tomography
density is often expressed in the form of CT Numbers or
Hounsfield Units.
From radiation intensity readings
acquired in the scanning process, the density or
attenuation values of tissues of interest can be
calculated.
The attenuation coefficients are normalized
74
with respect to water as the reference material, and a
magnifying constant is then applied to produce a CT
number.18
Hounsfield19 originally described the HU as an
absorption value; and he constructed a scale to demonstrate
the accuracy to which the absorption values could be
ascertained on a visual image.
For the machine he
described,19 the scale ranged from air (-1000) at the bottom
of the scale, to bone (1000) at the top of the scale.
Each
number represents a different shade of gray within the
spectrum.
The range of tones between black and white seen
in an image can be limited to a large or small window
within the scale.
This window can then be raised or
lowered depending upon the absorption value of the material
of interest.19
The examiner must be able to decide what
window level and width will most accurately represent the
anatomical truth of a tissue under examination.
It is
important for the clinician to be able to reliably detect
osseous abnormalities such as cortical erosion,
osteophytosis, osteoarthritis, and sclerosis.
The purpose
of the present study is to investigate what the ideal
Hounsfield unit values are for the examination of the
condyle; and if an ideal window width and level can be
identified, can one reliably evaluate the mandibular
condyle using the CBCT 3D reconstruction?
75
Materials and Methods
Selecting the Sample
The 25 dry human skulls in this study were used
with the permission of The Department of Anatomical
Science, Southern Illinois University at Edwardsville,
School of Dental Medicine.
No demographic data were
available on the skulls such as age, gender, or ethnicity.
Each of the 50 condyles was inspected for any signs of wear
or abrasion and documented photographically.
For each
condyle, six anatomic landmarks were identified, marked,
and photographed (see Figure 3.2).
Three linear measurements were made on the condyle
including: height, width, and length (see Table 3.2).
All
direct measurements were made by one operator using an
electronic digital caliper (P.N. 50001, Chicago Brand,
Fremont, CA) with orthodontic tips.
These measurements are
considered the “gold standard” (GS) for the purposes of
this study.
Reproduction of the original landmark
locations on the 3D renderings was assisted through the use
of photographs and markings on the condyles themselves.
76
SCo
ACo
MCo
PCo
LCo
a)
b)
L
c)
Figure 3.2. Anatomical landmarks: a) anterior view of
condyle showing medial mandibular condyle (MCo) and lateral
mandibular condyle (LCo), b) lateral view of condyle
showing posterior mandibular condyle (PCo), anterior
mandibular condyle (ACo), and superior mandibular condyle
(SCo), c) lingual view of the ramus showing lingula (L)
Imaging
CBCT scans of 25 dry human skulls were acquired
with the i-CAT cone beam CT scanner (Imaging Sciences
International, Hatfield, PA).
Skulls were mounted on a
polyvinylchloride stand that extended into the foramen
magnum providing support during the imaging process. A foam
wedge was placed between the glenoid fossa and the
capitulum of the condyle in order to separate the bony
77
surfaces; and the jaws were held together by bilateral
metal springs.
The skulls were oriented using the head
strap, and vertical and horizontal alignment lasers.
The
vertical laser was aligned with the midsagittal plane, and
the horizontal laser with the occlusal plane.
A second
vertical laser is aligned on the side of the skull just 1
inch anterior to the TMJ.
Lateral scout images were made
to check for any needed repositioning prior to making a
full scan.
The device was operated at 120 kVp and 3-8 mA
by using a high frequency generator with a fixed anode and
a 0.5mm focal spot.
A single 40 second high resolution
scan was made of each skull.
The dimensions of the field
of view (FOV) for each scan were 13cm for the height and
17cm for the diameter.
The voxel size was set at 0.25,
providing the greatest detail attainable with the i-CAT
CBCT scanner.
Primary data reconstruction was performed
immediately after data acquisition using the Imaging
Sciences i-CAT imaging software.
The DICOM (Digital
Imaging and Communications in medicine) data was burned to
a CD and transferred to a personal computer hard drive.
Multiplanar Reconstructions from the DICOM data were made
using V-works™4.0 imaging software by Cybermed Inc. (Seoul
Korea).
All operations were performed on a 17 inch flat
78
panel WXGA TFT active matrix display (Toshiba USA) with a
screen resolution of 1440 X 900, and operated at 32 bit.
After importing the DICOM data into V-works™4.0, proper
steps were taken to create images needed for the study.
To
do so, preset selection image options for the ‘head’ images
was selected, and the V-works™4.0 VR (Volume Rendering)
Preset ‘bone VR1’ was selected for the reconstruction.
Isolating and Measuring the Condyle
Each of the 50 condyles was isolated prior to
making the three-dimensional (3D) and volumetric
measurements.
Frankfort Horizontal was constructed by
creating a line from the inferior orbital rim to the
superior boarder of the external auditory meatus.
Using V-
works™4.0 volume operations and sculpting functions, an
initial cut was made parallel to Frankfort Horizontal just
above the superior aspect of the condyle, thereby removing
the majority of the surrounding bone.
The remaining
surroundings were progressively removed using various Vworks™4.0 sculpting tools (see Figure 3.3).
Three-dimensional multiplanar reconstructions were
produced for each of the 8 window widths defined in Table
3.1.
Reconstruction window width was set in the Volume
79
Table 3.1. Hounsfield Unit window widths (W) used to
create 3D renderings of the condyles. 3D linear
measurements were accomplished on each of these renderings.
Window Widths (W) for
3D Linear Measurements
W1
W2
W3
W4
W5
W6
W7
W8
-524
-424
-324
-224
-124
-24
76
176
to
to
to
to
to
to
to
to
2476
2476
2476
2476
2476
2476
2476
2476
mode using the histogram function.
HU
HU
HU
HU
HU
HU
HU
HU
Once this was
accomplished, it was exported as a SOD (Selection on
Demand); and a 3D model was created in the MPR mode.
All 3D reconstruction linear measurements were made
using the point-to-point measuring function in the SSD
(shaded surface display) mode.
The measurements include:
condylar width (CW), Condylar length (CL) and condylar
height (CH).
CW is defined as the linear distance between
the most lateral aspect of the lateral pole to the most
medial point on the medial pole.
CL is assessed by
measuring from the most posterior aspect of the condyle to
the most anterior point on the condylar head, and CH is
defined as the distance between the most superior point on
the condyle to the tip of the lingula.
These landmarks and
the measurements made using them are defined in Tables 3.2
80
a)
b)
c)
d)
Figure 3.3. 3D Reconstruction isolation: a) Initial lateral
view 3D reconstruction, b) Frankfort Horizontal initial
sculpting cut, c) Vertical sculpting cuts, d) Completed
isolation for condylar measurements.
81
and 3.3.
The 3D measurements made are shown in Figure 3.4.
In the event of elimination of portions of condylar
anatomy due to variation in window widths virtual planes
are constructed in the SSD mode.
Table 3.2.
Definitions of condylar linear measurements
Linear
distance
Measurement
Condylar
length (CL)
ACo – PCo
Condylar width
(CW)
MCo – LCo
Condylar
height (CH)
Table 3.3.
SCo - L
Definition
Linear distance between anterior
mandibular condyle and posterior
mandibular condyle
Linear distance between lateral mandibular
condyle and medial mandibular condyle
Linear distance between superior
mandibular condyle and lingula
Definitions of anatomic landmarks
Landmark
Anterior
mandibular condyle
(ACo)
Posterior
mandibular condyle
(PCo)
Definition
Most anterior extent of the mandibular condyle
viewed from the anterior, medial, lateral, and
superior planes of view.
Most posterior extent of the mandibular condyle
viewed from the posterior, medial, lateral, and
superior planes of view.
Lateral mandibular
condyle (LCo)
Most lateral extent of the mandibular condyle
viewed from the anterior, posterior, lateral, and
superior planes of view.
Medial mandibular
condyle (MCo)
Most medial extent of the mandibular condyle viewed
from the anterior, posterior, medial, and superior
planes of view.
Superior
mandibular condyle
(SCo)
Lingula (L)
Most superior aspect of the mandibular condyle
viewed from the anterior, posterior, medial,
lateral, and superior planes of view.
Apex of the lingula
82
b)
a)
c)
Figure 3.4. Point-to-point linear measurements on the 3D
reconstruction in V-works™4.0 SSD mode; a) Condylar width
(CW), b) Condylar length (CL), and c) Condylar height (CH).
83
These planes are constructed perpendicular to the line of
measurement.
As a defect increases the plane is advanced
into the body of the condyle to allow measurement of the
bony changes.
Measurements are made from the opposing
landmark to the constructed plane as demonstrated in
Figures 3.5 - 3.7.
Point-to-plane function was used for
these measurements.
Figure 3.5.
Lateral view of plane construction
84
Figure 3.6.
Figure 3.7.
measurement
Superior view of plane construction
Anterior view of plane construction with (CW)
85
Volume Measurements at Varying Window Widths
Prior to making volumetric measurements a complete
isolation of the condyle was completed.
Like the initial
slice, the final cut is made parallel to Frankfort
Horizontal at the level of the most inferior point in the
sigmoid notch.
The isolation process for volumetric
measurements is shown in Figure 3.8.
With this
accomplished, volumetric measurements were made for each of
the 22 window widths in order to define percentages of the
condylar volume within each window.
Each window represents
a range of bone densities defined in HU.
The first
volumetric measurement was made at W1 (176 HU to 275 HU),
and each of the remaining 22 volumetric measurements were
made in 99 HU width increments extending up to 2475 HU.
One final volumetric measurement was made with the total
range of 176 HU to 2475 HU (see Table 3.4).
All twenty-
four volumetric measurements were made in the V-works™4.0
multiplanar reconstruction mode.
Window width 24 was used
to find the total volume in the recommended bone density
range.
Figure 3.9 shows the volumetric 3D reconstructions
of the condyle at four different window widths.
86
a)
b)
c)
d)
Figure 3.8. 3D reconstruction isolation for volumetric
measurements of the mandibular condyle: a) initial cut
parallel to Frankfort Horizontal, b) second cut parallel to
the first cut at the level of the most inferior point in
the sigmoid notch, c & d) lateral and oblique views of the
final isolated reconstruction
87
a)
b)
c)
d)
Figure 3.9. Anterior and lateral volumetric renderings of
the right condyle at four different window widths: a) 1762476 HU, b) 376-475 HU, c) 1476-1575 HU, d) 1976-2075 HU
88
Table 3.4. Hounsfield Unit window widths (WW) used to
create 3D condyle renderings for volumetric evaluation.
Window 24 is equal to the sum of window widths 1-23.
WW1
176 to 275 HU
WW13
1376 to 1475 HU
WW2
276 to 375 HU
WW14
1476 to 1575 HU
WW3
376 to 475 HU
WW15
1576 to 1675 HU
WW4
476 to 575 HU
WW16
1676 to 1775 HU
WW5
576 to 675 HU
WW17
1776 to 1875 HU
WW6
676 to 775 HU
WW18
1876 to 1975 HU
WW7
776 to 875 HU
WW19
1976 to 2075 HU
WW8
876 to 975 HU
WW20
2076 to 2176 HU
WW9
976 to 1075 HU
WW21
2175 to 2275 HU
WW10
1076 to 1175 HU
WW22
2276 to 2375 HU
WW11
1176 to 1275 HU
WW23
2376 to 2475 HU
WW12
1276 to 1375 HU
WW24
176 to 2475 HU
Data Analysis
Differences between the 3D linear measurements and
the gold standard (GS) were calculated and analyzed using
SPSS 14.0 (SPSS Inc., Rainbow Technologies, Chicago, IL).
Significance testing for 3D linear measurement differences
was accomplished using independent t-tests with a 95%
confidence interval.
Linear measurement percent
differences were calculated in the 176 to 2476 HU window
for each of the fifty condyles.
individually analyzed.
CW, CL and CH were
The average percent linear
89
measurement change was calculated for the fifty condyles in
each measurement group.
All percent changes were plotted
on a distribution curve above and below the calculated
means for CW and CL.
The mean percent change was 18.38 for
CL and 15.93 for CW. The distribution of numbers was
segmented into the following groups: 1) numbers < 10% fall
into groups CW0 and CL0; 2) the mean group included numbers
in the 10-24% range; and 3) numbers >24% in groups CW1 and
CL1.
The mean group range of condyles was excluded from
the remainder of data analysis.
The distribution of
condyles in their perspective ranges are shown in Figure
3.10.
The percent volumes of the condyles in the outlying
groups were compared using independent t-tests at each of
the twenty-three volumetric window widths (WW) (see Table
3.5).
The CL0 and CL1 groups were plotted on a
distribution curve to compare the distribution of bone
volume in the two groups (see Figure 3.11).
The same was
done for the CW0 and CW1 groups on a separate graph (see
Figure 3.12).
Linear measurement error was calculated by
using the interclass correlation coefficient on the 12
repeat measurements for CL, CW, and CH.
90
Table 3.5. Hounsfield Unit window widths (WW) used to
create 3D condyle renderings for volumetric measurements
(see Table 3.4). CL1 represents the average % volume of
the condyles that showed the greatest change in condylar
length measurement at 176 HU – 2476 HU. CL0 was the
least change in condylar length measurement. CW1 was the
greatest change in condylar width, and CW0 the least.
Window Width
WW 1
WW 2
WW 3
WW 4
WW 5
WW 6
WW 7
WW 8
WW 9
WW 10
WW 11
WW 12
WW 13
WW 14
WW 15
WW 16
WW 17
WW 18
WW 19
WW 20
WW 21
WW 22
WW 23
CL1 (> 25%)
11.619
9.606
8.136
7.064
6.287
5.662
5.174
4.921
4.581
4.348
4.121
3.869
3.651
3.453
3.283
3.071
2.853
2.540
2.130
1.686
1.126
0.566
0.319
CL0 (< 10%)
9.928
8.678
7.612
6.726
5.986
5.323
4.876
4.522
4.184
4.009
3.863
3.750
3.682
3.541
3.517
3.535
3.528
3.322
2.913
2.450
1.930
1.339
0.788
CW1 (> 25%)
12.697
10.288
8.618
7.298
6.429
5.728
5.156
4.837
4.473
4.191
3.987
3.723
3.508
3.249
3.075
2.849
2.581
2.211
1.733
1.264
0.929
0.706
0.532
CW0 (< 10%)
9.468
8.414
7.730
6.872
6.108
5.428
4.962
4.626
4.281
4.143
4.021
3.909
3.864
3.757
3.726
3.759
3.804
3.655
3.352
2.921
2.262
1.423
0.747
Results
In the condyle sample (n = 50), 47 condyles were
intact.
The remaining three had minor abrasion defects
that were of no consequence to this study.
For 3D linear
measurement groups, significant differences were found
between the gold standards and CBCT measurements for the CL
91
90.00
85.00
80.00
% Linear Difference from GS
75.00
70.00
65.00
60.00
55.00
50.00
45.00
40.00
35.00
30.00
25.00
20.00
15.00
10.00
5.00
0.00
1
3
5
7
9
11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49
Condyles
a)
90.00
85.00
80.00
% Linear Difference From GS
75.00
70.00
65.00
60.00
55.00
50.00
45.00
40.00
35.00
30.00
25.00
20.00
15.00
10.00
5.00
0.00
1
3
5
7
9
11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49
Condyles
b)
Figure 3.10. Distribution of % Differences between 3D
linear measurements and gold standards (GS); a) CW
measurement % differences in W8 (176-2476 HU) with mean
group (yellow) between the red lines, b) CL measurement
differences in W8 with mean group (yellow) between the red
lines.
92
group at W7 and W8 (p < .05), and for the CW group at W6 (p
< .05), W7, and W8 (p < .01).
No significant measurement
differences were found for the CH group. The average linear
measurements for CL, CW, and CH along with the gold
standard (GS) are shown in Table 3.6.
With respect to the CL0 and CL1 groups, a
significant difference was found in percent volume at
windows 8, 9, 17, 18, and 19.
For the CW0 and CW1 groups
significant differences were found in WW 1-9 and 15-21.
Table 3.6. Hounsfield Unit window widths (W) used to
create 3D condyle renderings for linear measurements
(see Table 3.1 for W ranges). Average condylar length
(CL), condylar Width (CW), and condylar height (CH) are
compared with the gold standard (GS).
Comparison of CBCT mean linear measurements (mm) to gold standard
Linear
measurement
Windows
GS-CL
(mm)
CL
(mm)
Sig.
GS-CW
(mm)
CW (mm)
Sig.
GS-CH
(mm)
CH
(mm)
Sig.
W1
8.90
9.43
.058
18.50
19.10
.169
37.94
38.66
.590
W2
8.90
9.36
.098
18.50
18.90
.359
37.94
38.39
.736
W3
8.90
9.30
.149
18.50
18.65
.726
37.94
38.28
.800
W4
8.90
9.16
.342
18.50
18.24
.587
37.94
38.22
.832
W5
8.90
8.90
.988
18.50
17.67
.104
37.94
38.20
.849
W6
8.90
8.62
.442
18.50
16.78†
.004
37.94
38.15
.877
W7
8.90
8.08*
.040
18.50
15.66‡
.000
37.94
38.06
.927
W8
8.90
7.72†
.006
18.50
15.66‡
.000
37.94
37.62
.825
* p < .05; †P < .01; ‡P < .001
Table 3.7 lists independent t-test results and significance
for CL0 (n = 32) compared with CL1 (n = 15), and CW0 (n =
27) compared with CW1 (n = 16).
93
For both the CL and CW
groups, the mean bone percent volume distribution was
greatest in the low density windows and lowest in the high
density windows (see Figures 3.11 and 3.12).
CL0 and CW0
groups demonstrated greater percent volumes in the higher
density windows than the corresponding CL1 and CW1 groups.
Table 3.7. Comparison (Independent t-tests) of volumetric
measurements: CL0 vs. CL1, and CW0 vs. CW1 at twenty-three
window levels.
CL0 vs. CL1
Volumetric
Window
WW 1
WW 2
WW 3
WW 4
WW 5
WW 6
WW 7
WW 8
WW 9
WW 10
WW 11
WW 12
WW 13
WW 14
WW 15
WW 16
WW 17
WW 18
WW 19
WW 20
WW 21
WW 22
WW 23
CW0 vs. CW1
t
Sig.
t
Sig.
-1.682
-1.386
-1.152
-1.072
-1.309
-1.707
-1.626
-2.247
-2.358
-1.972
-1.597
-.742
.174
.435
1.044
1.812
2.323
2.323
2.091
1.861
2.038
2.288
1.543
.100
.173
.255
.289
.197
.095
.111
.030*
.023*
.055
.117
.462
.863
.665
.302
.077
.025*
.025*
.042*
.069
.047*
.027*
.130
3.417
2.925
2.637
2.165
2.489
2.697
2.265
2.296
2.211
1.197
.690
-.246
-1.154
-1.819
-2.350
-3.287
-4.210
-4.333
-4.403
-4.133
-3.292
-1.909
-.586
.001†
.006†
.012*
.036*
.017*
.010*
.029*
.027*
.033*
.238
.494
.807
.255
.076
.024*
.002†
.000‡
.000‡
.000‡
.000‡
.002†
.063
.561
* p < .05; †P < .01; ‡P < .001
94
The inverse relationship was evident in the lower density
windows.
So generally, the condyles that showed the
greatest linear measurement change in W8 (176 HU - 2476 HU)
were those with a higher percentage of low bone mineral
density (BMD); while those with little change in
measurement from the GS appear to have an increased BMD
content.
Linear measurement reliability was tested using the
intraclass correlation coefficient (ICC).
Repeat
measurements were accomplished on twelve condyles for all
three linear measurements.
A Cronboch’s Alpha of 0.917 was
found for the ICC test (ICC > .80 is acceptable).
Discussion
Studies have shown that CT images can be remarkably
accurate for linear,4,11,20,21 geometric,12 and volumetric22
measurements within the maxillofacial complex.
Hilgers et
al.4 found i-CAT CBCT images to be a nearly 1:1
reconstruction representation of the TMJ complex.
The
purpose to the present study was not to test the accuracy
linear measurements made on the 3D reconstruction, but
instead, to utilize its proven accuracy for the purposes of
measuring the changes that occur in the condyle as a result
95
CL1avg
CL0avg
12.000
10.000
Mean% Volume
8.000
6.000
4.000
* *
*
2.000
*
*
*
0.000
*
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
Window
*
P < .05;
†
< .01;
‡
p < .001
Figure 3.11. Distributions of condylar volume in twentythree window widths for the condyles of groups CL1 and CL0,
Window 1 represents the lowest density bone, and window 23
being the highest density bone observed.
of variation in reconstruction HU window level and window
width.
Window level and width variation for the 3D linear
measurements did have a significant effect on the condylar
width and length; however, changes in height were
statistically insignificant.
For this reason, CH was
excluded from the volumetric comparison groups.
CW was
most profoundly affected by window level and width.
96
Medial
CW1avg
CW0avg
12.500
Mean % Volume
10.000
†
7.500
†
*
*
*
5.000
*
*
*
*
*†
‡
2.500
‡
0.000
‡
‡
†
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
Window
*
P < .05;
†
< .01;
‡
p < .001
Figure 3.12. Distributions of condylar volume in twentythree window widths for the condyles of groups CW1 and CW0,
Window 1 represents the lowest density bone, and window 23
being the highest density bone observed.
and lateral poles of the 3D condylar reconstructions were
often the first to exhibit areas of erosion, and thereby
producing reduced CW measurements.
Significance was found
in the following ranges: -24 to 2476 HU, 76 to 2476 HU, and
176 to 2476 HU.
The highest density window level, 176 to
2476 HU, is the recommended window width for viewing bone
97
with V-works™4.0.
to measure.
CL proved to be a challenging dimension
ICC reliability testing showed measurement
reproducibility to be acceptable; however, the extent of
erosions was difficult to measure with point to point, and
or point to plane measurements.
This likely resulted in
fewer windows with significant measurement differences.
So
the CL was possibly more profoundly affected by changes in
window level and width than what the results present.
The Hounsfield Unit has been used to describe
physical density and achieve reasonable volume estimates of
anatomic structures.13-15,18,22,23
The greater a tissue
density, the higher the HU numbers will be within the
respective window width.
A number of studies have been
performed in efforts to quantify bone density based on
Hounsfield values.13-15
Many of these studies were
accomplished in order to help classify bone types best
suitable to support dental implants.13,14,17
For the present
study, HU window widths and levels are manipulated in order
to create 3D reconstructions most representative of the
anatomic truth.
For CW and CL the most accurate windows
were below the recommended window for bone, and extended
into the soft tissue range as defined in Table 3.8.
In a
dry skull, extending into the soft tissue range will
enhance visualization.
However, in vivo the soft tissue
98
Table 3.8. Window widths for density values and Hounsfield
numbers in the body.24
will begin to appear and diminish ones capacity to view the
bony topography. This would suggest that, by itself, the
CBCT 3D reconstructed image may not be a reliable way to
diagnose condylar pathology and or changes in condylar
morphology.
Though there was significant measurement change in
the in the 176 to 2476 HU window for CL and CW groups,
there was a range of variation within each group for the
fifty condyles.
Some condyles exhibited large change while
others showed little or no change at all.
It is important
to define why these condyles are different from each other.
Is an increased difference in linear measurements from the
anatomical truth directly related to the BMD composition of
the condyle?
The purpose of defining the volumetric
distribution of BMD within the condyles was intended to
99
answer this very question.
The two groups compared
graphically in Figures 3.11 and 3.12 show the volumetric
distribution over 23 density ranges.
Significant
differences in bone density distribution were found for
both the CL and CW groups.
Windows of significant
difference were in the high and low density ranges.
No
significant differences were found in the midrange windows
(WW 10 to WW 14).
The number of significant windows was
greater for the CW0/CW1 group than the CL0/CL1 group.
One
would expect the CW and CL density distributions to be
alike.
They do follow the same pattern; however, the CL
demonstrated fewer windows of significance between the CL0
and CL1 groups.
This could relate to the difficulty
presented in measuring morphologic defects in this
dimension.
Incapacity to accurately record the anatomic
changes occurring across the range of HU windows could
produce error in condyle selection for the CL0 and CL1
groups; thereby, producing a volumetric comparison between
two groups of condyles with very little variation between
them.
It was not the goal of this study to define the
range of bone mineral densities that constitute the
condyle.
A study of this nature would require a sample of
untreated cadaver condyles.
Even then, research has shown
100
that, HU are only reliable as BMD predictors in full
trabecular bone, or in the presence of minimal cortical
bone.
HU readings from CT scans with thicker cortical bone
become less reliable.15,16
Also, CT numbers can not be
accepted as an absolute for characterization of a tissue
type or lesion.25
CT numbers may vary significantly from
one scanner to another, or even between two scanners of the
same make and model.25,26
However, with some degree of
success, standardized calibration methods have been
employed in efforts to minimize inter-scan discrepancies.27
With attention to detail, strict standardization in all
parameters, continual manufacturer support, and application
of proper calibration methods, reproducibility can be
optimized.28
Studies focused on CT numbers as an accurate
representation of tissue density have, for the most part,
been confined to conventional CT scanners.
Recently,
Aranyarachkul et al.13 examined variations in bone density
in designated implant recipient sites using both CBCT and
conventional CT.
They found both modalities to be
consistent in their measurements of bone density value;
however, the values were generally higher for CBCT.
Whether CBCT or conventional CT values are closer to
corresponding histological bone densities has yet to be
investigated.
101
Further research on the application of CBCT imaging
in the examination of the TMJ is warranted.
One such study
would be to investigate the accuracy and reliability of
CBCT in producing density values that correspond with
histological bone densities.
Defining the range and
distribution of expected bone densities within the condyle
may assist the clinician in diagnosis morphologic changes
and pathologic processes within the TMJ.
On the medical
side, CT values have already been applied, with some degree
of success, in the differentiation of adrenal masses.29
Nwariaku et al.29 found that a threshold attenuation value
of ≤ 10 HU accurately characterized adrenal cortical
adenomas.
This concept could also be applied to the
maxillofacial complex in the evaluation of both hard and
soft tissues.
Conclusions
Due to the relatively high composition of low
density bone within the mandibular condyle, radiographic
assessment, using the 3D reconstruction, is most accurate
when accomplished at density levels below that recommended
for osseous examination.
However, utilization of lower
window levels, extending into the soft tissue range, may
compromise ones capacity to view bony topography.
102
Depending on the distribution of bone density within the
condyle, CBCT 3D reconstructed images, by themselves, may
not be a reliable means for diagnosis of condylar pathology
and or changes in condylar morphology.
One might, instead,
employ the 3D reconstruction simply as a screening tool,
and perform comprehensive condylar examinations using
progressive crossectional imaging.
Using the multiplanar
reconstruction viewing mode is an excellent way to
accomplish this.
103
Acknowledgements
In acknowledgement of the people who have helped on
this thesis project.
I would like to thank Dr. Ki Beom Kim
for his constant support and dedication through the entire
process.
I would also like to thank Drs. Donald Oliver and
Gus Sotiropoulos for devoting their time, knowledge, and
encouragement along the way.
Also, this project would not
have been possible without the help of Dr. Cyrus Alizadeh,
Dr. Becky Schreiner and the Alizadeh Orthodontics staff.
Thank you for your gracious hospitality and selfless
contribution to help make this project a success. Finally,
thanks to Drs. Heidi Israel and Binh Tran for their help
with the statistical analysis of the data.
104
Literature Cited
1. Brooks SL, Brand JW, Gibbs SJ, Hollender L, Lurie AG,
Omnell KA et al. Imaging of the temporomandibular joint: a
position paper of the American Academy of Oral and
Maxillofacial Radiology. Oral Surg Oral Med Oral Pathol
Oral Radiol Endod 1997;83:609-618.
2. Quintero JC, Trosien A, Hatcher D, Kapila S.
Craniofacial imaging in orthodontics: historical
perspective, current status, and future developments. Angle
Orthod 1999;69:491-506.
3. Sukovic P. Cone beam computed tomography in craniofacial
imaging. Orthod Craniofac Res 2003;6 Suppl 1:31-36;
discussion 179-182.
4. Hilgers ML, Scarfe WC, Scheetz JP, Farman AG. Accuracy
of linear temporomandibular joint measurements with cone
beam computed tomography and digital cephalometric
radiography. Am J Orthod Dentofacial Orthop 2005;128:803811.
5. Mah JK, Danforth RA, Bumann A, Hatcher D. Radiation
absorbed in maxillofacial imaging with a new dental
computed tomography device. Oral Surg Oral Med Oral Pathol
Oral Radiol Endod 2003;96:508-513.
6. Honda K, Larheim TA, Maruhashi K, Matsumoto K, Iwai K.
Osseous abnormalities of the mandibular condyle: diagnostic
reliability of cone beam computed tomography compared with
helical computed tomography based on an autopsy material.
Dentomaxillofac Radiol 2006;35:152-157.
7. Kau CH, Richmond S, Palomo JM, Hans MG. Threedimensional cone beam computerized tomography in
orthodontics. J Orthod 2005;32:282-293.
8. Hatcher DC, Aboudara CL. Diagnosis goes digital. Am J
Orthod Dentofacial Orthop 2004;125:512-515.
105
9. Ludlow JB, Davies-Ludlow LE, Brooks SL, Howerton WB.
Dosimetry of 3 CBCT devices for oral and maxillofacial
radiology: CB Mercuray, NewTom 3G and i-CAT.
Dentomaxillofac Radiol 2006;35:219-226.
10. Walker L, Enciso R, Mah J. Three-dimensional
localization of maxillary canines with cone-beam computed
tomography. Am J Orthod Dentofacial Orthop 2005;128:418423.
11. Lascala CA, Panella J, Marques MM. Analysis of the
accuracy of linear measurements obtained by cone beam
computed tomography (CBCT-NewTom). Dentomaxillofac Radiol
2004;33:291-294.
12. Marmulla R, Wortche R, Muhling J, Hassfeld S. Geometric
accuracy of the NewTom 9000 Cone Beam CT. Dentomaxillofac
Radiol 2005;34:28-31.
13. Aranyarachkul P, Caruso J, Gantes B, Schulz E, Riggs M,
Dus I et al. Bone density assessments of dental implant
sites: 2. Quantitative cone-beam computerized tomography.
Int J Oral Maxillofac Implants 2005;20:416-424.
14. Shapurian T, Damoulis PD, Reiser GM, Griffin TJ, Rand
WM. Quantitative evaluation of bone density using the
Hounsfield index. Int J Oral Maxillofac Implants
2006;21:290-297.
15. Stoppie N, Pattijn V, Van Cleynenbreugel T, Wevers M,
Vander Sloten J, Ignace N. Structural and radiological
parameters for the characterization of jawbone. Clin Oral
Implants Res 2006;17:124-133.
16. Rho JY, Hobatho MC, Ashman RB. Relations of mechanical
properties to density and CT numbers in human bone. Med Eng
Phys 1995;17:347-355.
17. Hatcher DC, Dial C, Mayorga C. Cone beam CT for presurgical assessment of implant sites. J Calif Dent Assoc
2003;31:825-833.
106
18. Mull RT. Mass estimates by computed tomography:
physical density from CT numbers. AJR Am J Roentgenol
1984;143:1101-1104.
19. Hounsfield GN. Nobel Award address. Computed medical
imaging. Med Phys 1980;7:283-290.
20. Kobayashi K, Shimoda S, Nakagawa Y, Yamamoto A.
Accuracy in measurement of distance using limited cone-beam
computerized tomography. Int J Oral Maxillofac Implants
2004;19:228-231.
21. Misch KA, Yi ES, Sarment DP. Accuracy of cone beam
computed tomography for periodontal defect measurements. J
Periodontol 2006;77:1261-1266.
22. Breiman RS, Beck JW, Korobkin M, Glenny R, Akwari OE,
Heaston DK et al. Volume determinations using computed
tomography. AJR Am J Roentgenol 1982;138:329-333.
23. Kobayashi F, Ito J, Hayashi T, Maeda T. A study of
volumetric visualization and quantitative evaluation of
bone trabeculae in helical CT. Dentomaxillofac Radiol
2003;32:181-185.
24. Cybermed. V-works™4.0. Seoul, Korea: Cybermed, co, Ltd;
2006: p. Imaging software for real time 3D visualization.
25. Levi C, Gray JE, McCullough EC, Hattery RR. The
unreliability of CT numbers as absolute values. AJR Am J
Roentgenol 1982;139:443-447.
26. Stadler A, Schima W, Prager G, Homolka P, Heinz G,
Saini S et al. CT density measurements for characterization
of adrenal tumors ex vivo: variability among three CT
scanners. AJR Am J Roentgenol 2004;182:671-675.
27. Cann CE. Quantitative CT for determination of bone
mineral density: a review. Radiology 1988;166:509-522.
107
28. Cann CE. Quantitative CT applications: comparison of
current scanners. Radiology 1987;162:257-261.
29. Nwariaku FE, Champine J, Kim LT, Burkey S, O'Keefe G,
Snyder WH, 3rd. Radiologic characterization of adrenal
masses: the role of computed tomography--derived
attenuation values. Surgery 2001;130:1068-1071.
108
Vita Auctoris
Brian Albert Schlueter was born on July 12, 1974 to
Robert R. and Marsha A. Schlueter at Wiesbaden Air Force
Base in Wiesbaden, Germany.
Shortly after, the Schlueter
family moved back to the states, and Brian and his two
siblings Scott and Amanda were raised in St. Louis,
Missouri.
He attended high school at Westminster Christian
Academy from 1989-1993.
Following graduation, Brian
attended Texas Christian University in Fort Worth, TX where
he studied biology and chemistry.
Upon graduation from TCU, Brian was married to
Hollie Marie Anderson; and began his education in dentistry
at SIUE School of Dental Medicine in Alton, IL.
After
graduating from SIUE and receiving his degree (DMD), they
moved to Travis Air Force Base, CA where Brian began his
active duty service in the USAF, and attended an advanced
general dentistry residency.
While in California, Brian
and Hollie had their first son Noah Grant.
Upon completion
of his active duty commitment to the USAF in 2004 at
Whiteman AFB, MO, Brian and his family moved home to St.
Louis, MO to began his orthodontic residency training at
Saint Louis University.
On July 19th 2006 they became a
109
family of four with the birth of their second son Parker
Gabriel.
After finishing his orthodontic residency, Brian,
Hollie, Noah and Parker plan to remain in the St. Louis
area, where the rest of their family resides.
110