The purposes of this study were to determine and compare the intraexaminer and interexaminer reliabilities of commonly used cephalometric landmarks identified on digitized lateral cephalograms and formatted cone-beam computerized tomography (CBCT) images.
CBCT images from 10 randomly selected adolescent patients were obtained from the orthodontic records of a private practice. Measurement errors, and intraexaminer, and interexaminer reliability correlation coefficients (ICC) were obtained for all landmark coordinates.
Intraexaminer and interexaminer reliabilities for all coordinates for most landmarks on the digital lateral cephalograms and CBCT images were greater than 0.9 (ICC value). The means of landmark locations differed by approximately 1 mm in most coordinates from the lateral cephalograms and were predominantly higher than 1 mm for all coordinates from the CBCT images.
Intraexaminer and interexaminer reliabilities were high for most landmarks. Coordinates with greater measurement errors in the lateral cephalograms (condylion, gonion, porion, mandibular incisor apex, and posterior nasal spine) were in structures without clearly defined borders. In the CBCT images, gonion, condylion, and porion were located on surfaces that were flat or curved, making it difficult to recognize a specific reference point. Other less reliable landmarks (anterior nasal spine, posterior nasal spine, mandibular incisor apex) were located in structures with lower densities and could not be visualized with 3-dimensional reconstruction; thus, they had high measurement errors.
Since the development of cephalometric radiology, several cephalometric analyses have been proposed. They have been useful in describing how individual patients vary from population norms, forecasting and following growth and treatment changes, and establishing descriptive communications between clinicians. Because cephalometric analysis is a 2-dimensional (2D) rendering from 3-dimensional (3D) structures, cephalometric measurements on radiographic images are subject to projection, landmark identification, and measurement errors.
Magnification and distortion play important roles in the radiographic projection errors of skeletal and dental structures in cephalometric images. Magnification occurs because the x-ray beams originate from a source that is not parallel to all points of the object examined. Distortion occurs because of unequal magnifications between different planes. Although many landmarks used in cephalometric analysis are located in the midsagittal plane and are not prone to superimposition errors, other landmarks with different paramedial structures are affected by distortion because of their locations at different depth fields.
Landmark identification errors are also considered a major source of cephalometric errors. This type of error is influenced by many factors such as quality of the radiographic image, precision of landmark definition, reproducibility of the landmark location, and operator and registration procedures. Despite all these potential errors, cephalometric radiographs are still widely used and are often essential in a patient’s diagnosis and treatment.
Advances in the use of 3D imaging hardware and software have challenged our perception of 3D craniofacial structures and their associated growth. Monitoring of treatment changes is also affected. CBCT is a relatively new technique that allows primary reconstructions (sagittal, coronal, and para-axial cuts) and secondary reconstructions (3D reconstructions and maximum intensity projections) of various craniofacial structures. Compared with traditional cephalometric radiographs, CBCT images are anatomically true (1:1 in size) 3D representations from which slices can be displayed from any angle in any part of the skull and provided digitally on paper or film.
Currently, 3D volumetric imaging provides useful information for clinicians in identifying teeth and other structures for diagnostic and descriptive purposes. Before establishing CBCT as a common orthodontic diagnostic approach, landmark reliability must be assessed. This has been extensively done for traditional lateral cephalograms. However, landmark reliability assessment for CBCT is limited, and additional research is required in this area. The purposes of this study were to determine and compare intraexaminer and interexaminer reliabilities of common cephalometric landmarks from digitized lateral cephalograms and formatted 3D CBCT images.
Material and methods
Digitized lateral cephalograms (Planmeca, Roselle, Ill) and CBCT scans (NewTom 3G volumetric scanner, Aperio Services, Verona, Italy) from 10 adolescent patients were randomly selected from the orthodontic records previously taken at a private practice orthodontic clinic in Calgary, Alberta, Canada. The sample size was based on a statistical power of 0.90 with α = 0.05. This study was approved by the Human Research Ethics Board at the University of Alberta.
After obtaining the CBCT images (by using a 12-in field of view with 8-mm aluminum filtration at 110 kV and 6.19 mAs, and slice thickness of 0.5 mm) in raw study data, they were converted into DICOM format. Commercially available third-party software (AMIRA, Mercury Computer Systems, Berlin, Germany) was used to obtain primary reconstructed images (axial, coronal, and sagittal) and the 3D reconstructions of the images for landmark recognition and location. Lateral cephalograms (obtained at 68 kV and 12 mA, and image size with approximately a 12-in field of view) were uploaded into the AMIRA software, and landmark locations were calculated.
AMIRA software has a predetermined fiduciary coordinate axis system for each image. The center of the coordinate axis system is outside the image of interest. This predetermined coordinate axis system is always the same when the same image is uploaded in the software. Since the purpose of this study was not to compare images, determining a common reference plane on every image was not necessary.
The landmarks used in this study are described in Table I . For the coordinates obtained from CBCT, the AMIRA software gave values in millimeters. The CBCT data had no magnification (1:1 image size), and, to allow true comparison, magnification of the lateral cephalogram images was corrected with the calibration ruler imbedded in each image at its acquisition.
|Nasion (N): most anterior point of the frontonasal suture in the median plane|
|Orbitale (Or): lowest point in the inferior margin of the orbit|
|A-point (A): point at the deepest midline concavity on the maxilla between ANS and prosthion|
|B-point (B): point at the deepest midline concavity on the mandibular symphysis between infradentale and pogonion|
|Pogonion (Pg): most anterior point of the bony chin in the median plane|
|Gnathion (Gn): most anteroinferior point on the symphysis of the chin, constructed by intersecting a line drawn perpendicular to the line connecting menton and pogonion|
|Menton (Me): most inferior midline point on the mandibular symphysis|
|Gonion (Go): constructed point of intersection of the ramus plane and the mandibular plane|
|Porion (Po): superior point of the external auditory meatus|
|Sella (S): midpoint of the pituitary fossa (sella turcica)|
|Basion (Ba): median point of the anterior margin of the foramen magnum|
|Anterior nasal spine (ANS): tip of the anterior nasal spine|
|Posterior nasal spine (PNS): tip of the posterior nasal spine|
|Condylion (Co): most superior point on the condylar head|
|Upper central incisor tip (U1T): point on the tip of the maxillary central incisor crown|
|Upper central incisor root apex (U1R): point on the apex of the maxillary central incisor root|
|Lower central incisor tip (L1T): point on the tip of the mandibular central incisor crown|
|Lower central incisor root apex (L1R): point on the apex of the mandibular central incisor root|
Landmark coordinates for each image set were obtained by 1 investigator (M.O.L.) 3 times, and 1 time by 2 investigators (C.F. and R.C.). All examiners were previously trained in the use of AMIRA software and orthodontic landmark identification. For investigator blinding, the images were identified by code and analyzed in random order. Intraexaminer reliability was assessed by using intraclass correlation coefficients (ICC) for the first investigator’s 3 measurements. ICC was also used to calculate interexaminer reliability by comparing his second trial with the measurements of the other 2 investigators. Measurement errors (average of the mean differences between measurement trials) for all coordinates (x, y, and z for CBCT; x and y for digital lateral cephalograms) were also determined.
Intraexaminer and interexaminer reliabilities for the x and y coordinates of most landmarks in the lateral cephalograms were greater than 0.9. Only porion, basion, and condylion had moderate intraexaminer reliabilty for the y-axis (0.81, 0.57, and 0.67, respectively) and mild interexaminer reliability for the y-axis (0.46, 0.46, and 0.38, respectively).
Mean differences from repeated landmark identification by the same examiner in the x-axis were less than 1 mm with the exception of posterior nasal spine (1.52 mm) and condylion (1.38 mm). For the y-axis, the mean differences were equal to or less than 1 mm with the exception of basion (1.64 mm), condylion (1.36 mm), and mandibular incisor root apex (1.23 mm). When the 3 examiners were compared, their mean differences in the x-axis were less than 1 mm in 50% of the landmarks, with gonion (2.81 mm), basion (1.46 mm), anterior nasal spine (1.58 mm), maxillary incisor root apex (1.66 mm), mandibular incisor root apex (1.38 mm), and posterior nasal spine (2.26 mm) all greater than 1 mm. In the y-axis, the greatest differences were in gonion (2.28 mm), basion (2.45 mm), porion (1.96 mm), condylion (2.12 mm), maxillary incisor root apex (2.59 mm), and mandibular incisor apex (2.36 mm) ( Tables II and III ) .
Intraexaminer and interexaminer reliabilities for the x, y, and z coordinates for all landmarks in CBCT were greater than 0.9.
Mean differences from the same examiner’s trials were generally less than 1.0 mm. In the x-axis, orbitale left, sella, basion, anterior nasal spine, posterior nasal spine, and condylion right had values between 1.0 and 2.0 mm. Porion right and left had the highest differences in this axis (2.62 and 3.37 mm, respectively). In the y-axis, gonion right and left, porion left, and posterior nasal spine had mean differences between 1.0 and 2.0 mm. In the z-axis, only B-point and mandibular incisor root apex left had mean differences between 1.0 and 2.0 mm.
For the mean differences between the 3 examiners in the x-axis, they were predominantly higher than 1.0 mm. Orbitale right and left (3.25 and 2.57 mm, respectively), porion right and left (2.7 and 2.94 mm, respectively), and condylion right and left (3.48 and 3.08 mm, respectively) all had mean differences greater than 2.0 mm. In the y-axis, half of the landmarks had errors higher than 1.0 mm. Gonion right and left (5.5 and 3.9 mm, respectively) and anterior nasal spine (2.51 mm) had mean differences greater than 2.0 mm. In the z-axis, about 40% of the landmarks had errors higher than 1.0 mm. Gonion right and left (3.5 and 2.66 mm, respectively) and mandibular incisor root apex left (2.05 mm) all had mean differences greater than 2.0 mm ( Tables IV and V ) .