Surg Radiol Anat (2002) 24: 393–399 DOI 10.1007/s00276-002-0058-x
R AD IO L OG I C AN A TO M Y
H.-J. Kim Æ H.-R. Yoon Æ K.-D. Kim Æ M.-K. Kang H.-H. Kwak Æ H.-D. Park Æ S.-H. Han Æ C.-S. Park
Personal-computer-based three-dimensional reconstruction and simulation of maxillary sinus Received: 1 February 2002 / Accepted: 6 July 2002 / Published online: 21 January 2003 Ó Springer-Verlag 2003
Abstract Anatomical descriptions of the maxillary sinus are critical in pathological diagnosis and the treatment planning of surgical procedures. This study was undertaken to develop a new technique for simulating anatomical structures and to clarify the morphological and clinical characteristics of the maxillary sinus. Thirtythree hemi-sectioned Korean heads were used in this study. CT scans and DentaScan reformatted cross-sectional images were taken on all specimens. From the CT images, three-dimensional reconstructed images were made using the V-works program. From the three-dimensional reconstructed images of the maxillary sinus, six categories of maxillary sinus were created, categorized according to their lateral aspects and shapes of the inferior walls. In 55%, a flat inferior wall of the maxillary sinus was observed. All measurements (anteriorposterior length, height, width and volume) of the sinus
were larger in males than in females. From the DentaScan reformatted panoramic images, the anterior limit of the maxillary sinus was located in the first premolar area (58%), and the posterior limit was in the third molar and maxillary tuberosity area (94%). We therefore offer a new virtual technique for manipulating three-dimensional reconstructed images easily on a personal computer. On the reconstructed images the threedimensional morphology could be observed and the anatomical characteristics of the maxillary sinus and surrounding structures could be determined. The French version of this article is available in the form of electronic supplementary material and can be obtained by using the Springer Link server located at http:// dx.doi.org/10.1007/s00276-002-0058-x.
Reconstruction tridimensionnelle sur ordinateur personnel et simulation du sinus maxillaire The French version of this article is available in the form of electronic supplementary material and can be obtained by using the Springer Link server located at http://dx.doi.org/10.1007/s00276002-0058-x H.-J. Kim (&) Æ M.-K. Kang Æ H.-H. Kwak Æ H.-D. Park Division in Anatomy, Department of Oral Biology, College of Dentistry, Oral Science Research Center, Brain Korea 21 Project for Medical Science, Yonsei University, 134 Shinchon-Dong, Seodaemoon-Gu, Seoul, 120-752, Korea E-mail:
[email protected] Tel.: +82-2-3618044 Fax: +82-2-3938076 H.-R. Yoon Æ K.-D. Kim Æ C.-S. Park Department of Oral and Maxillofacial Radiology, College of Dentistry, Yonsei University, Seoul, Korea S.-H. Han Catholic Institute of Applied Anatomy, Department of Anatomy, College of Medicine, Catholic University, Seoul, Korea
Re´sume´ Les e´tudes anatomiques du sinus maxillaire sont importantes pour le diagnostic de sa pathologie et la planification de son abord chirurgical. Ce travail a e´te´ re´alise´ afin de de´velopper une nouvelle technique de simulation des structures anatomiques et pour pre´ciser ainsi les caracte´ristiques morphologiques et cliniques du sinus maxillaire. 33 he´mi-sections de teˆtes de sujets core´ens ont e´te´ utilise´es. Des tomodensitome´tries et des reconstructions avec le logiciel DentaScan ont e´te´ re´alise´es pour tous les spe´cimens. A partir des images tomodensitome´triques, des reconstructions tridimensionnelles ont e´te´ effectue´es avec un logiciel V-works. Graˆce a` ces images tridimensionnelles, une classification des sinus maxillaires en six cate´gories, selon la forme de leur paroi infe´rieure e´tudie´e en vue late´rale, a e´te´ e´tablie. Dans 55% des cas, la paroi infe´rieure du sinus e´tait plate. Toutes les mesures (longueur ante´ro-poste´rieure, hauteur, largeur et volume) du sinus e´taient plus e´leve´es chez l’homme que chez la femme. Sur les reformations panoramiques obtenues par le DentaScan, la limite ante´rieure du sinus maxillaire
394
e´tait le plus souvent situe´e en regard de la premie`re pre´molaire (58%), et sa limite poste´rieure en regard de la troisie`me molaire et de la tube´rosite´ maxillaire (94%). Avec tous ces e´le´ments, nous proposons une nouvelle technique virtuelle de reconstruction et d’exploitation d’images tridimensionnelles, d’utilisation facile sur un ordinateur personnel. Graˆce a` ces images, la configuration tridimensionnelle et les caracte´ristiques anatomiques du sinus maxillaire et des structures avoisinantes ont pu eˆtre e´tudie´es. Keywords Maxillary sinus Æ Inferior wall of maxillary sinus Æ Maxillary teeth Æ Computed tomography Æ DentaScan reformatted image
Introduction Conventional dental radiographs have been used for the evaluation of anatomical structures and pathoses of the maxillary sinus. However, it is difficult to identify small bony defects and the relationships between the inferior wall of the sinus and the root apex using these conventional radiographs [11, 14, 17]. Functional three-dimensional visualization of detailed anatomical structures is being developed in accord with the requirements of current surgical procedures. Despite the accuracy of present imaging modalities for the evaluation of anatomical structures, some limitations with respect to detailed morphology and spatial relationships of oral and maxillofacial structures remain in clinical dentistry, especially in terms of dental implantation procedures. Nowadays, DentaScan reformatted cross-sectional computed tomography (CT) images can provide an evaluation of the complexity of anatomical structures from initial diagnosis to final treatment. Since reformatted cross-sectional images were first applied in the clinical field in the 1980s [12], three-dimensional radiographic images have been used to evaluate and plan treatment for various head and neck pathoses [6, 7, 21], oral and maxillofacial trauma [8], congenital malformation and reconstructive surgery [2, 5]. Nowadays, CT applications are increasing in the clinical dentistry field [1, 9, 10, 16], especially anthropological studies on the measurements of the head and neck by CT [7, 13]. However, often several related specialists working together are required for a completely accurate operation. Therefore, functional three-dimensional visualization approaches on a personal computer system will be useful to clinicians prior to undertaking surgical procedures. The aim of this study was to develop a new technique for simulating the maxillary sinus structures based on a three-dimensional simulation system (V-works 3.0 program). In addition, using these three-dimensional images of the maxillary sinus we attempted to clarify the morphological and clinical characteristics of the inferior sinus wall.
Materials and methods Thirty-three sides of maxillae from hemi-sectioned Korean heads (19 males, 14 females; average age 55.8 years) were used in this study. Computed tomography (GE Medical Systems, Milwaukee, Wis., USA) was performed on all specimens. To obtain the reformatted CT images, a CT HiSpeed Advantage (GE Medical Systems, Milwaukee, Wis., USA) with a highresolution bone algorithm, 15 cm field of view, 200 mA, 120 kV, scanning time of 1 s and slice thickness of 1 mm was used to obtain the axial images from the occlusal margin of the maxillary teeth to the inferior margin of the orbit. The axial images were reformatted using DentaScan software (GE Medical Systems, Milwaukee, Wis., USA), and the reformatted cross-sectional images were used for the radiographic evaluation of the inferior wall of maxillary sinus and to observe the anterior and posterior limit of the sinus (Fig. 4). Using the CT data as Digital Imaging and Communications in Medicine (DICOM) files, three-dimensional reconstructed images of the maxillary sinuses were made using the V-works (version 3.0) program (CyberMed, Seoul, Korea, http://www.cybermed.co.kr) (Fig. 1). The basic functions of V-works are as follows: (1) The system can reconstruct three-dimensional models of human organs from various data sources, such as CT and MRI. (2) Users can manipulate the models using a variety of interactive functions. For example, a user can browse a model using a mouse, navigate inside the model, or examine a cross-sectional view by registering and overlapping a 2D image on a threedimensional model. (3) Using the V-works simulation module, a user can create operation plans using a marking, measurement, cutting, and movement functions. We performed various measurements on three-dimensional reconstructed images using V-works 3.0 program. The measurement parameters used were the anterior-posterior maximum length, height, width and volume of the sinus. We classified the three-dimensional reconstructed images of the maxillary sinuses into six categories, based on the gross morphology of the lateral aspect and the shapes of the inferior sinus wall (Fig. 2).
Results Three-dimensional reconstruction and morphometry of the maxillary sinus We classified 33 three-dimensional reconstructed images, in V-works 3.0 program, into six categories based on the morphology of the inferior wall of the maxillary sinus (types I–VI). Type I, with a flat wall at the premolar and molar areas, was observed in eight cases (24%). In types II and III the inferior wall was narrower than the superior wall, and also showed a flat inferior maxillary sinus wall in the maxillary molar region in seven cases (type II, 21%) and a slanted inferior wall in the maxillary premolar area in five cases (type III, 15%). Types IV and V had a round and acutely angled inferior wall at the maxillary second premolar and first molar area in seven cases (21%) and three cases (9%), respectively. Cases of type VI showed a wider maxillary sinus inferior wall than superior wall in three cases (9%) (Table 1, Fig. 2).
395 Fig. 1 Frame images captured by the V-works 3.0 program
To identify the lowest level of the inferior wall of the maxillary sinus, we observed the three-dimensional reconstructed images and DentaScan cross-sectional radiographic images. In nine cases the lowest levels of the inferior wall of sinus were located at the maxillary first molar. In the others, these were located at the interproximal area between the maxillary second and third molars (6 cases), at the second molar (6 cases), at the third molar (4 cases), at the interproximal area between
the second premolar and the first molar (2 cases), and between the first and second molars (2 cases) (Fig. 3). From the three-dimensional reconstructed images of the maxillary sinus, the maximum anteroposterior length of the sinus was 39.3±4.2 mm (male: 40.7 mm, female: 37.4 mm), its maximum height was 37.1±5.6 mm (male: 39.4 mm, female: 34.0 mm) and maximum width was 32.6±6.5 mm (male: 35.3 mm, female 28.9 mm). The average volume of the sinuses was
396
Fig. 2 Photographs and figures showing the lateral views of the three-dimensional reconstructed maxilla and maxillary sinus (red colored region) on the V-works 3.0 program
Table 1 Classifications of the three-dimensional morphology of the maxillary sinus based on the morphology of the inferior wall in the lateral view (numbers represent the number of samples observed) Type
Males (n=19)
Females (n=14)
Total (%)
I II III IV V VI
5 4 2 3 2 3
3 3 3 4 1 0
8 7 5 7 3 3
(24%) (21%) (15%) (21%) (9%) (9%)
15.1±6.2 ml. All measurements were larger in males than females (Table 2). Morphology of the maxillary sinus on DentaScan reformatted panoramic images Using 24 sides of DentaScan reformatted panoramic images (Fig. 4) we observed the anterior and posterior limit of the maxillary sinus. In 58% of cases the anterior limit of the maxillary sinus was located at the first premolar area (14 sides), in 33% at the canine area (8 sides) and in 8% at the second premolar area (2 sides). Most posterior limits of the maxillary sinus were located at the third molar and maxillary tuberosity area (22 cases, 94%); the others were at the maxillary second molar area (2 cases, 6%).
Discussion As the medical and dental fields become more specialized, the need for precise diagnosis of anatomical and pathological structures increases. To achieve this diagnostically, radiographic technologies have advanced following the development of CT and MRI in the 1970s. Nowadays, multiplanar reformatted images and threedimensional radiographic images are used in various fields, and these newly developed imaging techniques can be used to identify detailed structures in the human body. Anatomically, the maxillary sinus is an air-filled space in the maxilla. It has a pyramidal shape and is composed of four walls. It is well known that the morphology and size of the maxillary sinus vary [18, 22]. In particular, it has been reported that maxillary sinus volume correlates with the interzygomatic buttress distance [4]. However, factors of malocclusion, state of the dentition and gender have no influence on the size of the maxillary sinus, but gender is a significant factor only in Angle’s class II malocclusion [3, 20]. For these reasons, we did not consider the sinus dimensions according to the dentition state and malocclusion in this study. Due to anatomical complexities, it is difficult to identify three-dimensional human structures using conventional radiographic techniques, in particular the evaluation of the relationship between the inferior wall of the sinus and the apices of the maxillary teeth. Moreover, cross-sectional images have proven to be very useful in diagnosis. For this reason, CT images are today essential in the diagnosis of maxillary sinus pathoses.
397
Fig. 3A–F The lowest levels of the inferior wall of the maxillary sinus on three-dimensional reconstructed images. The position of the lowest level of the maxillary sinus is shown at the maxillary first molar (A), at the interproximal area between the maxillary second and third molars (B), at the second molar (C), at the third molar (D), at the interproximal area between the second premolar and the first molar (E), and between the first and second molars (F)
two-dimensional axial images from CT to three-dimensional images using volume-rendering techniques. In this study, we created three-dimensional images of the maxillae and maxillary sinuses from two-dimensional axial images using the V-works 3.0 program. Using these three-dimensional reconstructed images, we undertook some measurements and morphological
Recently, multiplanar reformatted images have proven to be very useful for the identification of threedimensional structures and the spatial relationships of pathoses in the oral and maxillofacial structures. Moreover, the uses of multiplanar reformatted images are increasing. However, these reformatted images have been obtained from professional workstation computer systems that are directly connected with the CT apparatus. Thus, a three-dimensional image reconstruction program operating on a personal computer system can be useful due to its cost effectiveness. The V-works 3.0 program used in this study can easily convert the
Table 2 Measurements of the maxillary sinus using the three-dimensional reconstructed images on the V-works program (values are mean±SD)
Maximum A-P length (mm) Maximum height (mm) Maximum width (mm) Volume (ml)
Males (n=19)
Females (n=14)
Total (n=33)
40.7±4.5
37.4±3.0
39.3±4.2
39.4±5.8
34.0±3.5
37.1±5.6
35.3±6.9 18.0±6.2
28.9±3.5 11.1±3.4
32.6±6.5 15.1±6.2
Fig. 4 DentaScan image showing the reformatted panoramic views of the maxillary teeth and maxillary sinus. At the left side of the upper panel, the anterior and posterior limit of the maxillary sinus is located at the first premolar and second molar area, respectively. In the lower panel, the most posterior limit of the maxillary sinus is located at the maxillary tuberosity area. 4, first premolar; 5, second premolar; 6, first molar; 7, second molar; 8, third molar
398
classifications based on the shape of the inferior wall of the maxillary sinus. From the lateral aspect, the morphology of the maxillary sinus of Koreans was classified into six categories. Among these classifications, the categories described as having flat inferior walls of the maxillary sinus (types I and II) were most prominent in Koreans (45%). Interestingly, cases that showed a wider inferior wall than superior wall of the maxillary sinus (type VI) were observed only in male specimens (9%), showing a gender difference (Table 1, Fig. 2). Morphometric measurements of the reconstructed three-dimensional images using V-works identified such gender differences. All measurements had a tendency to be greater in males than in females in terms of sinus volume (Table 2). However, the material was limited in the present study, which may make it difficult to resolve this question clearly. In Gray’s Anatomy [22], the overall measurements of the maxillary sinus are described as 32 mm in anterior-posterior diameter (length), 25 mm in mediolateral diameter (width), and 35 mm in superior-inferior diameter (height). Compared with the Korean data, it appears that all measurements in Koreans are greater than have been reported previously [22]. On the DentaScan reformatted panoramic images, the anterior and posterior limits of the maxillary sinus can easily be identified. Generally, the anterior limit of the maxillary sinus of the Japanese is located at the second premolar (44.8%), first molar (31.1%) and first premolar region (19.0%), and the posterior limit lies at the third molar (48.3%), second molar (27.6%) and maxillary tuberosity region (24.1%) [15]. However, our results differed, as in the present study the anterior limit of the maxillary sinus was located at the maxillary first premolar region in 58% of cases, the limit in the other cases being at the canine (33%) and second premolar region (8%). Moreover, the posterior limit of the maxillary sinus was located at the third molar and the maxillary tuberosity region in 94% of cases. It is known that the inferior wall of the maxillary sinus is proximate to the first or second molar root apex [23] and that the lowest level of the maxillary sinus is located at the maxillary second molar region [19]. However, the present study clarifies that the lowest level of the inferior wall of the maxillary sinus in Koreans was located at the maxillary first molar area, the interproximal region between the second and third molar and the second molar region in turn, which differed from previous reports.
Conclusion We have conducted a new virtual technique for reconstructing three-dimensional images that are easily manipulated on a personal computer system. On these reconstructed images, the three-dimensional morphology could be easily observed and the anatomical
characteristics of the maxillary sinus and surrounding structures visualized. Moreover, we provide an application of this reformatted image technique in the area of oral and maxillofacial diagnosis and treatment planning. Acknowledgements This work was supported by the BK21 Project for Medical Science, Yonsei University. We would thank NamKuk Kim, CEO of Cybermed Co., Young-Im Kim, Jae-Bum Lee for their effort in this simulation system development.
References 1. Abrahams JJ, Berger SB (1998) Inflammatory disease of the jaw: appearance on reformatted CT scans. Am J Radiol 170: 1085–1091 2. Altobelli DE, Kikins R, Mulliken JB, Cline H, Lorensen W, Jolesz F (1993) Computer-assisted three dimensional planning in craniofacial surgery. Plast Reconstr Surg 92: 576–585 3. Ariji Y, Kuroki T, Moriguchi S, Ariji E, Kanda S (1994) Age changes in the volume of the human maxillary sinus: a study using computed tomography. Dentomaxillofac Radiol 23: 163– 168 4. Ariji Y, Ariji E, Yoshiura K, Kanda S (1996) Computed tomographic indices for maxillary sinus size in comparison with the sinus volume. Dentomaxillofac Radiol 25: 19–24 5. Carls FR, Schuknecht B, Sailer HF (1994) Value of three-dimensional computed tomography in craniofacial surgery. J Craniofac Surg 5: 281–285 6. Cavalcanti MGP, Vannier MW (1998) Quantitative analysis of spiral computed tomography for craniofacial clinical applications. Dentomaxillofac Radiol 27: 344–350 7. Cavalcanti MGP, Vannier MW (1998) The role of three-dimensional spiral computed tomography in oral metastases. Dentomaxillofac Radiol 27: 203–209 8. Fox L, Vannier MW, West CO, Wilson JA, Baran GA, Pilgram TK (1995) Diagnostic performance of CT, MPR, 3DCT imaging in maxillofacial trauma. Comput Med Imaging Graphics 19: 385–395 9. Fuhrmann R, Bu¨ker A, Diedrich PR (1997) Furcation involvement: comparison of dental radiographs and HR-CTslices in human specimens. J Periodontal Res 32: 409–418 10. Fuhrmann R, Bu¨ker A, Diedrich PR (1997) Radiological assessment of artificial bone defects in the floor of the maxillary sinus. Dentomaxillofac Radiol 26: 112–116 11. Gutteridge DL (1995) The use of radiographic techniques in the diagnosis and management of periodontal disease. Dentomaxillofac Radiol 24: 107–113 12. Herman GT, Coin CG (1980) The use of three-dimensional computer display in the study of disk disease. J Comput Assist Tomogr 4: 564–567 13. Hildervolt CF, Vannier MW, Knapp RH (1990) Validation study of skull three-dimensional computerized tomography measurements. Am J Phys Anthropol 82: 283–294 14. Jeffcoat MK (1992) Radiographic methods for the detection of progressive alveolar bone loss. J Periodontol 63: 367–372 15. Kamijo M (1990) Oral anatomy, part 1: Osteology, 2nd edn. Anatom, Tokyo, pp 209–214 (in Japanese) 16. King JM, Caldarelli DD, Petasnick JP (1992) DentaScan: a new diagnostic method for evaluating mandibular and maxillary pathology. Laryngoscope 102: 379–387 17. Lang VP, Hill RW (1977) Radiographs in periodontics. J Clin Periodontol 4: 16–28 18. McGowan DA, Baxter PW, James J (1993) The maxillary sinus and its dental implications, 1st edn. Wright, London, pp 1–25 19. Mustian WF (1933) The floor of the maxillary sinus and its dental and nasal relation. J Am Dent Assoc 20: 2175–2187
399 20. Oktay H (1992) The study of the maxillary sinus areas in different orthodontic malocclusions. Am J Orthod Dentofac 102: 143–145 21. Ray CE, Mafee MF, Friedman M, Tahmoressi CN (1993) Applications of three-dimensional CT imaging in head and neck pathology. Radiol Clin North Am 31: 181–194
22. Williams PL, Bannister LH, Berry MM, Collins P, Dyson M, Dussek JE, Ferguson MWJ (1995) Gray’s anatomy, 38th edn. Churchill Livingstone, Edinburgh, p 1637 23. Worth HM (1963) Principles and practice of oral radiographic interpretations. Year Book Medical Publishers, Chicago, pp 42–52, 697–716
