427

CLINICS IN PLASTIC SURGERY

Clin Plastic Surg 34 (2007) 427436

Three-Dimensional Computerized Orthognathic Surgical Treatment Planning

cia V.M. Alves, Patr Linping Zhao, PhDc
-

DDS, MS

a,b,

*, Ana M. Bolognese,

DDS, MS, DSc

Acquisition of the three-dimensional medical images Determination of landmarks and cephalometric analysis

Three-dimensional computerized virtual treatment objective Case study Summary References

Orthognathic surgery involves anatomically dening the deformity, establishing an appropriate orthodontic-surgical treatment plan, and then executing the recommended treatment. The surgical-orthodontic team must then not only predict the various possible outcomes based on the options available, but once agreed, must execute that plan as precisely as possible. Historically, the surgical-orthodontic planning has relied on two-dimensional (2D) analysis of radiographic images, the lateral cephalometric lm. Currently available cephalometric software planning uses standard osteotomies (the LeFort I, bilateral sagittal split osteotomy [BSSO], and genioplasty) and soft tissue prediction based on 2D analysis to develop a virtual treatment plan. For the surgeon, however, a 2D blueprint using only the sagittal plane as a guide for executing a three-dimensional (3D) surgical procedure in multiple planes is less than ideal. Moreover, standard osteotomies in such software packages cannot simulate asymmetric osteotomies, or the

osteotomies that are more appropriately tailored to correct the specic deformities, such as a modied LeFort I, in which to varying degrees the zygoma are included. Therefore, each of the elements of the craniofacial structure may require a complex osteotomy pattern and individual 3D manipulation to achieve an optimal outcome. In recent years, 3D CT imaging with reconstruction of 2D-acquired data has become routinely available in the clinical setting. Compared with conventional 2D dentofacial lms, the 3D images provide signicantly better visualization of the morphology. Thus as new tools are developed, we can navigate away from the limitations of conventional 2D cephalometry with 3D CT [1,2]. Today, however, use of the 3D imaging data for orthognathic surgical planning remains in the setting of clinical research, because there are few accepted standards or conventions for managing such 3D computational data and methodology for surgical planning [36]. In this article, the authors discuss

Federal University of Rio de Janeiro, Brazil University of Illinois at Chicago, Chicago, IL, USA c Shriners Hospitals for Children-Chicago, 2211 North Oak Park Avenue, Chicago, IL 60707, USA * Corresponding author. 2018 West Adams, Chicago, IL 60612. E-mail address: [email protected] (P.V.M. Alves).
b

0094-1298/07/$ see front matter 2007 Elsevier Inc. All rights reserved.

doi:10.1016/j.cps.2007.04.006

plasticsurgery.theclinics.com

428

Alves et al

our current approach to a 3D computerized visual surgical planning system.

Acquisition of the three-dimensional medical images

Three-dimensionally reconstructed CT has been available for a number of years; however, it has been primarily used for diagnosis and visualization of the morphologic anatomy in three dimensions rather than formal surgical planning. When it is needed, a 3D solid model is generated from the CT data, and the surgery is performed on a physical model. Solid model surgery is not routine, however, because it is limited by an additional expense, and once sectioned, it requires another model to assess another variation. Ultimately such an approach is limited, in that soft tissue prediction cannot be assessed. Thus virtual surgery on a data set has inherent advantages over a solid model, in that it allows the surgeon an unlimited number of potential outcomes and the possibility of predicting soft tissue response with each possible variation. The critical step will then be translating the virtual model surgery to intraoperative execution with the use of surgiguides or intraoperative navigation tools. Additionally, at this time we have limited data to program the soft tissue response in 3D. Recent advances in CT scan technique have brought about reductions in both cost and exposure to radiation [6]. The application of multidetector helical CT scanner is increasing. New CT scanners with multiple detectors (16, 64, and even 128 detectors) can considerably reduce the time of scanning, and therefore reduce the artifacts caused by body movement. These machines allow the acquisition of 3D images with smaller slice thickness within acceptable radiation exposure [7]. The cone beam CT scan can perform a full scan of the head in around 20 seconds, and gives the patient an effective dose of only 50 mSv, compared with about 2000 mSv from a typical conventional CT scan of the whole head [8]. Cone-beam CT produces a lower radiation dose than spiral CT, and is almost comparable to panoramic radiographs. It also allows secondary reconstructions, such as sagittal, coronal, and axial views, and 3D reconstructions of various craniofacial structures because of its volumetric data [9,10]. Beyond this statement, radiation concerns are further reduced when one considers that a single CT scan replaces a number of conventional radiographs that are now considered essential for almost every orthognathic patient, such as the lateral cephalogram, occlusal, periapical, and panoramic radiograph [2,11]. The resolution of the soft tissue, as muscles, in cone beam

CT scan is not as good as in the helical CT scan, however. This soft tissue information needs to be obtained from MRI. With the limitation of the clinical acceptable radiation exposure, the spatial resolution of multidetector helical CT scan can reach the order of 0.3 to 0.5 mm, with a slice thickness of 0.5 to 1.0 mm. The spatial resolution of cone beam CT scan can reach the order of 0.3 to 0.5 mm. This brings anatomical and functional information with sufcient accuracy not only for diagnosis, but also for virtual craniomaxillofacial surgery planning and navigation. Three-dimensional cephalometry using standardized CT scanning protocols has the advantage that all measurements are life-sized scale (1:1), which allows both cross-sectional and longitudinal comparison of 3D distances, linear projective, and orthogonal measurements. Unlike conventional cephalometric radiography, spiral CT-based 3D cephalometry does not necessitate standardized xation of the skull during record taking, because the 3D virtual scene approach allows standardized virtual positioning of the skull to the Frankfort horizontal plane [11]. Compared with the traditional cephalometric radiographs, cone-beam computerized tomography (CBCT) also produces images that are anatomically true (1:1 in scale). Three-dimensional representations from the slices can be displayed from any angle in any part of the skull [1214]. Advances in both computer hardware and image analysis software now enable interactive display of the data on personal computers, with the ability to selectively view soft or hard tissues from any angle. Both in plane and 3D, measurement tools allow clinicians to extract useful information conveniently. Technological progress makes it possible to produce accurate 3D anatomical virtual models, which is important for cephalometric analysis and surgical planning. Using medical image analysis software, one can build 3D virtual models from a set of several axial cross-sectional slices, and perform the 3D computerized virtual treatment objective. In this article, the authors demonstrate our approach to virtual surgical treatment planning using Mimics version 10.11 (Materialise Belgium, Leuven, Belgium). The use of the medical image analysis software starts with importing image data from CT and MRI. This is a critical step. Several issues must be considered. The scanned data should be reconstructed as axial slices, with one to one ratio so as to preserve all of the information. The data format of the CT/MRI scan should be standardized as a Digital Imaging and Communications in Medicine

Three-Dimensional Orthognathic Surgical Treatment

429

(DICOM) le. The data transfer can be performed via a CD or DVD disc, but online systems such as Picture Archiving and Communication Systems (PACS) are preferred. The selection from a wealth of information should start from the data import step, but not before. The next step is the segmentation of anatomical structures, which is the process of outlining the shape of structures visible in the cross-sections of a volumetric data set. Based upon the Hounseld scale, hard and soft tissue can be separated; even the structurelike airway can be dened. The voxelbased manipulation denes both the surface and volume of an object, and allows further separation of tissues and removal of artifacts that need to be cleaned (a voxel is a volume element, representing a value on a regular grid in 3D space. This is analogous to a pixel, which represents the smallest unit of 2D image data). A 3D graphic rendering of the volumetric object allows navigation between voxels in the volumetric image and the 3D graphics with zoom, rotate, and pan.

Determination of landmarks and cephalometric analysis

Despite the availability of 3D CT images for a number of years, 3D cephalometric studies for the assessment of dentofacial skeletal abnormalities are still at an early stage [1521]. This is largely because of the lack of a practical 3D landmark system that allows the cephalometric analysis to be performed, and compared with the conventional 2D cephalometric database that has been in existence for nearly a half-century. The authors have developed a 3D cephalometric landmark system integrated into the software that we have found useful for surgical planning. The landmarks and 3D cephalometric measurements are based on a well-accepted anatomical landmark system, and tailored to t to our application in orthognathic surgery and treatment. Accurate identication of landmarks requires anatomic knowledge and experience in landmark denition. Compared with conventional cephalometric radiography, some landmarks (eg, anterior nasal spine) are easier, whereas others (eg, sella) more difcult to dene in 3D cephalometry. Moreover, sometimes the denition of conventional 2D cephalometric landmarks has to be modied because of the third dimension, or new 3D cephalometric landmarks (eg, posterior maxillary point) have to be dened for computing 3D cephalometric planes. Three-dimensional landmark identication, however, requires suitable operational denitions of the landmark location in each of the three planes

of space (X, Y, and Z). When a point is indicated on the cephalogram image generated from CT scan, the 3D point should be positioned on the bone surface. Therefore, the geometrical relationship between cephalogram and CT image volume is a prerequisite if one is to benet from the combination of CT and virtual cephalograms. This combination of 2D and 3D information is the key to accurate indication of landmarks in a repeatable way. Landmark identication errors are considered the major source of cephalometric error. This type of error is inuenced by many factors, such as the quality of the radiographic image, the precision of landmark denition, the reproducibility of the landmark location, the operator, and the registration procedure. Three-dimensional cephalometry provides highly accurate measurements of lengths and angles dened by landmarks directly placed on the surface of the bones. The landmarks should be used as a center point or registration plane for the 3D cephalometry coordinate system in preoperative diagnosis and surgical planning references. Also, landmarks at the cranial base, foramina, canals, and sutures can be used as guides. Ono and colleagues [22] reported high accuracy with less than 3% measurement errors using a prototype of helical CT. Other studies for cephalometric measurements have yielded equivalent accuracy [23,24]. A 3D cephalometric system enables better description of bilateral points as Porion (Po), Orbitale (Or), Gonion (Go), and others. This allows for a more accurate assessment of the asymmetries involved when planning the surgical correction [2527].

Three-dimensional computerized virtual treatment objective

Once the anatomical landmark system has been established, medical image analysis software such as SimPlant CMF and Mimics (Materialise Belgium) can generate the projection on the midsagittal plane, and this can be compared with conventional 2D cephalometric analysis and ideal values. To have a more accurate comparison, such software should allow the operator to redene the midsagittal plane according to the landmarks. It should also allow the reorientation of the images, and generate the 2D image slices along any chosen plane. The use of software cutting tools allows segmentation of bone, and varying osteotomies can be simulated. The position tools allow the translation and rotation of skeletal elements in 3D space once the bones are segmented. The software also has the capacity to simulate the soft tissue response, and allows the superimposition of soft tissue images over the skeletal framework. Thus various surgical options can be assessed to achieve an optimal

430

Alves et al

outcome. One of the advantages of 3D virtual surgery planning is that multiple procedures can be simulated and evaluated. Without limitations of the cephalometric radiograph, such as representation of objects that is symmetrical to the midsagittal plane, 3D virtual surgery allows the surgeon freedom in simulating a complex surgical procedures that represent a combination of different osteotomies with skeletal structural repositioning in all three planes.

Case study
This is best illustrated through a case study. This patient presented with a at prole and short upper lip (Fig. 1). The 2D lateral cephalometric demonstrates dentofacial deformity characterized by antero posterior relation of Class III with a corresponding Class III molar relation (Fig. 2). The 3D cephalometric conrms these features, and also shows the midline asymmetry, posterior crossbite, and the landmark system to perform the diagnosis and treatment plan (Fig. 3). The landmarks nomenclature used for our protocol is based on the recommendations found in the 4th edition of Nomina Anatomica [28]. These anatomical terminologies are internationally accepted as anthropometric symbols. The main difference between the 2D and 3D is that 3D analysis allows better evaluation of

symmetry, by measuring bilateral anatomic points from sagittal plane, instead of a superimposing, as with a 2D analysis. Table 1 exemplies a 2D and a 3D cephalometric analysis, as well as the comparison between them. Virtual treatment objectives are built by using the Mimics software. The surgical options to achieve skeletal relation of Class I and improve function and esthetics range from a mandibular setback alone to a midfacial skeletal advancement of a LeFort I type at various levels (Fig. 4). After orthodontic preparation, the surgical procedure was planned to sagittally advance the maxilla, vertically to increase dental display and a rotation from right to left. This planning resulted in a modied LeFort I (Fig. 5). The nal results are shown by clinical photographs (Fig. 6), 2D (Fig. 7) and 3D records (Fig. 8). There is close correlation between the soft tissue prediction and the nal result at 1 year (Fig. 9) (Table 1). Fig. 9 shows the superimposition between the postsurgical CT scan and the nal model generated from virtual treatment objective, as well as the 3D models from preoperative and postoperative CT scan.

Summary
Three-dimensional volumetric imaging allows better visualization of the morphologic deformity,

Fig. 1. Presurgical records: (A) facial frontal view, (B) 45 prole view right, (C) prole view right, (D) intra oral prole right view, (E) frontal view, and (F) prole view left.

Three-Dimensional Orthognathic Surgical Treatment

431

Fig. 2. Conventional orthognathic surgery planning utilizes 2D lateral cephalometric radigraphs and analysis (A). The surgery is then executed using facebow transfer (B) and mounted dental casts (C) to perform the model surgery (D) based on the 2D cephalometric analysis.

Fig. 3. Three-dimensional representation from CT scan shows the anatomic structures and cephalometric landmarks before surgery. (AC) Skull and hyoid bone. (DF) Soft tissue, spine, and airway added. Presurgical cephalometric analysis based on 2D (Dolphin software) and 3D (Mimics software) is described in Table 1.

432

Alves et al

Fig. 4. Three-dimensional virtual treatment objective showing different possible surgical options: (A) Mandibular setback with a bilateral sagittal split ramal osteotomy, (BD) variations of the LeFort I procedure, (E) dentofacial skeletal deformity characterized by complex asymmetries in multiple planes, and (F) virtual model surgery planning showing the osteotomy planes, simulation of the movements of the jaws, and possible bone graft.

Three-Dimensional Orthognathic Surgical Treatment

433

Table 1: Comparison between 2D cephalometric measures before/after treatment; and 3D cephalometric measures before orthodontic/orthognathic treatment, virtual treatment objective (VTO), and after orthodontic/orthognathic treatment 2D analysis Pre SNA SNB ANB U1NA IMPA Orl-Sg plane Orr-Sg plane U6l-Sg plane U6r-Sg plane Pr-Sg plane Id-Sg plane 86 89 -3 128 75


3D analysis Pre 85 88 -3 140 84 31 mm 29 mm 15 mm 12.5 mm 0.76 mm 3.17 mm



Post 90 88 2 133 72



VTO 88.2 87 1.20 143 84 31.05 mm 29.50 mm 16.05 mm 11.66 mm 2.12 mm 2.78 mm


Post 88.5 87 1.5 145 80 31 mm 29 mm 16 mm 12 mm 1.90 mm 3.33 mm

SNA, Sella Nasion A point angle; SNB, Sella Nasion B point angle; ANB, A point Nasion B point angle; U1NA, upper central incisor and Nasion A point angle; IMPA, lower central incisor and mandibular plane angle; Orl-Sg plane, linear measure from Orbitale left to sagital plane; Orr-Sg plane, Orbitale right to sagital plane; U6l-Sg plane, palatal gengival border of the upper left molar to sagital plane; U6r-Sg plane, upper right molar to sagital plane; Pr-Sg plane, Prosthion point to sagital plane; Id-Sg plane, Infradentale point to sagital plane.

planning the surgical approach, and evaluating the response not visualized previously with 2D dentofacial records. This in time will replace conventional 2D cephalometric planning and plaster cast model surgery. Three-dimensional craniofacial imaging requires application of various techniques from applied mathematics, computer sciences,

and bioengineering. Although today in its infancy, it has the potential to accurately simulate the operative experience for surgical planning, with improved morphologic outcomes, patient-specic biomechanical modeling to allow functional assessment of the various outcomes, and accurate simulation for surgical resident training.

Fig. 5. Surgical treatment plan based on the patients CT scan and 3D cephalometrics. The maxilla is advanced in the sagittal plane with vertical lengthening. Additionally, a right-to-left rotation to correct the asymmetry is planned. This replaces conventional model surgery illustrated in Fig. 2.

434

Alves et al

Fig. 6. Postsurgical records: (A) facial frontal view, (B) 45 prole view right, (C) prole right view, (D) intra oral prole right view, (E) frontal view, and (F) prole left view. (Surgeon P. K. Patel, MD; Orthodontist Kevin W. Ensley, DDS.)

Fig. 7. Preoperative (A) and postoperative (B) 2D lateral cephalometric radiographs. The analysis in described in Table 1.

Three-Dimensional Orthognathic Surgical Treatment

435

Fig. 8. Three-dimensional CT scan following modied LeFort I advancement with cephalometric landmarks. AC show skull and hyoid bone. DF have soft tissue, spine and airway added. The 3D analysis is described in Table 1.

Fig. 9. (A) Preoperative and postoperative soft tissue comparison based on actual (not predicted) CT data with the modied LeFort I advancement. (B) Soft tissue prediction based on the modied LeFort I (Fig 4) and postoperative CT scan at 1 year, showing good predication with slight underestimate. The soft-tissue prediction will be expected to improve with renement in the mathematic algorithm with accumulation of 3D outcomes data over time.

436

Alves et al

References
[1] Wolford LM, Karras SC, Mehra P. Cosiderations for orthognathic surgery during growth. Part 1: Mandibular deformities. Am J Orthod Dentofacial Orthop 2001;119(2):95101. [2] Halazonetis DJ. From 2-dimensional cephalograms to 3-dimensional computed tomography scans. Am J Orthod Dentofacial Orthop 2005; 127(5):62737. [3] Elolf E, Tatagiba M, Samii M. Three dimensional computed tomographic reconstruction: planning tool for surgery of skull base pathologies. Comput Aided Surg 1998;3(2):8994. [4] Franca C, Levin-Plotnik D, Sehgal V, et al. Use of three-dimensional spiral computed tomography imaging for staging and surgical planning of head and neck cancer. J Digit Imaging 2000; 13(2 Suppl 1):2432. [5] Cavalcanti MG, Vannier MW. Measurement of the volume of oral tumors by three-dimensional spiral computed tomography. Dentomaxillofac Radiol 2000;29(1):3540. [6] Swennen GRJ, Schutyser F, Hausamen JE. Threedimensional cephalometry. A color atlas and manual. Berlin: Springer; 2006. p. 364. [7] Ngan DC, Kharbanda OP, Geently JP, et al. Comparison of radiation levels from computed tomography and conventional dental radiographs. Aust Orthod J 2003;19(2):6775. [8] Mah JK, Danforth RA, Bumann A, et al. Radiation absorbed in maxillofacial imaging with a new dental computed tomography device. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2003;96(4):50813. [9] Ziegler CM, Woertche R, Brief J, et al. Clinical indications for digital volume tomography in oral and maxillofacial surgery. Dentomaxillofac Radiol 2002;31(2):12630. [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(4):41823. [11] Cevidanes LHS, Styner MA, Proft WR. Image analysis and superimposition of 3-dimensional cone-beam computed tomography models. Am J Orthod Dentofacial Orthop 2006;129(5): 6118. [12] Maki K, Inou N, Takanishi A, et al. Computerassisted similations in orthodontic diagnosis and the application of a new cone beam x-ray computed tomography. Orthod Craniofac Res 2003;(6 Suppl 1):95101. [13] Mah J, Hatcher D. Current status and future needs in craniofacial imaging. Orthod Craniofac Res 2003;(6 Suppl 1):106. [14] Farman AG. Fundamentals of image acquisition and processing in the digital era. Orthod Craniofac Res 2003;(6 Suppl 1):1722.

[15] Hatcher DC, Aboudara CL. Diagnosis goes digital. Am J Orthod Dentofacial Orthop 2004; 125(4):5125. [16] Hajeer MJ, Ayoub AF, Millett DT, et al. Three-dimensional imaging in orthodontic surgery: the clinical application of a new method. Int J Adult Orthod Surg 2002;17(4):31830. [17] Chirani RA, Jacq JJ, Meriot P, et al. Temporomandibular joint: a methodology of magnetic resonance imaging 3-D reconstruction. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2004;97(6):75661. [18] Nkenke EN, Zachow S, Benz M, et al. Fusion of computed tomography data and optical 3D images of the dentition for streak artifact correction in the simulation of orthognathic surgery. Dentomaxillofac Radiol 2004;33(4):22632. [19] Harrell WE Jr, Hatcher DC, Bolt RL. In search of anatomic truth: 3-dimensional digital modeling and the future of orthodontics. Am J Orthod Dentofacial Orthop 2002;122(3):32530. [20] Miller AJ, Koutaro M, Hatcher DC. New diagnostic tools in orthodontics. Am J Orthod Dentofacial Orthop 2004;126(4):3956. [21] Xia J, Samman N, Yeung RWK, et al. Three-dimensional virtual reality surgical planning and simulation workbench for orthognathic surgery. Int J Adult Orthodon Orthognath Surg 2000; 15(4):26582. [22] Ono I, Gunji H, Suda K, et al. Method of preparing an exact-size model using helical volume scan computed tomography. Plast Reconstr Surg 1994;93(7):136371. ` re MO, Major PW. Proposed reference point [23] Lagrave for 3-dimensional cephalometric analysis with cone-beam computerized tomography. Am J Orthod Dentofacial Orthop 2005;128(5):65760. ` re MO, Hansen L, Winfried H, et al. Plane [24] Lagrave orientation for standardization in 3-dimensional cephalometric analysis with computerized tomography imaging. Am J Orthod Dentofacial Orthop 2006;129(5):6014. [25] Kitaura H, Yonetsu K, Kitamori H, et al. Standardization of 3-D CT measurements for length and angles by matrix transformation in the 3-D coordinate system. Cleft Palate Craniofac J 2000;37(4):34956. [26] Kim N-K, Lee C, Kang S-H, et al. Chang Y-II. A three-dimensional analysis of soft and hard tissue changes after a mandibular setback surgery. Comput Methods Programs Biomed 2006;83(3): 17887. [27] Hajeer MY, Ayoub AF, Millett DT. Three-dimensional assessment of facial soft-tissue asymmetry before and after orthognathic surgery. Br J Oral Maxillofac Surg 2004;42(5):396404. [28] Federative Committee on Anatomical Terminology. Terminologia Anatomica. New York: Thieme, 1998.