Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
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