This article addresses the use of CT and other imaging modalities in oral and maxillofacial surgery, with an emphasis on dentoalveolar surgery. In the 2005 issue of the Atlas of the Oral and Maxillofacial Surgery Clinics of North America , Dodson provided an excellent overview on “Interactive CT Software in Oral and Maxillofacial Surgery.” This article provides an update on the use of CT and presents additional information that reflects further changes and advances that will be of use to the practicing surgeon. We encourage the reader to read the article by Dodson in conjunction with ours.
Obtaining the CT scan
Before one can use a scan, one needs to obtain a scan. Not all scans are alike. There is such a thing as a bad scan, or at least not an optimal scan. Practitioners should have a clear understanding of the differences between the two major groups of scanners, and of the technical parameters that should be used for a specific scan. This knowledge is important even if a practitioner refers a patient to an imaging center, but is particularly significant for practitioners who have an in-office scanner.
The two major groups of scanners are conventional or medical CT and cone beam CT (CBCT). Conventional CT scanners have been in clinical use since the mid-1970s . The scanners have several differences, but two are most clinically relevant. One difference is that CBCT offers a significantly reduced radiation dose compared with medical CT, which is the single most important reason why CT has become widely used in dentistry. The other difference is that CBCT does not offer the same soft tissue contrast as conventional CT. For most dental and oral surgical procedures, the lack of soft tissue contrast is not a problem because dentists are mostly concerned with hard tissues. However, when definition of muscle is desirable, versus structures such as fat, conventional CT would be the better study. Of course, if the primary interest is the soft tissues, then MRI would be the preferred modality.
Differences also exist among CBCT scanners. They are broadly divided into large, medium, and small field-of-view (FOV; the latter sometimes called limited FOV ) machines. As their names indicate, large FOV machines image a larger area. In general, the larger the FOV, the poorer the resolution. Small FOV CBCTs are especially suited to endodontics, in which high resolution is needed to visualize canals, root fractures, endodontic fill, and other tooth-related anomalies. A single implant site or two sites can also be imaged with a small FOV machine, but because the resolution required for implant planning is not nearly as great as for endodontic purposes, one should reduce the resolution for an implant study to keep the radiation dose as low as reasonably achievable. If, as seems likely, an oral surgeon wishes to use the CBCT to image pathology not related or limited to just a tooth or two, then a large FOV machine is probably needed, because the small FOV units do not cover a sufficiently large area. Although two or more small FOV images can be “stitched” together, this is probably not practical for a busy oral and maxillofacial practice to do on a routine basis.
Even this short description of CBCT units should make it evident that no “one size fits all” machine exists. Surgeon wanting to acquire a machine would benefit greatly from consulting with an oral and maxillofacial radiologist or someone who is knowledgeable about the various machines. Additionally, a Web site is available that provides comparative information on the CBCT machines on the market ( http://www.conebeam.com/cbctchart ), and is updated periodically.
Practitioners should follow a few basic guidelines to obtain an optimal scan. Scatter from metallic objects is an inherent shortcoming of all CT scans, both medical and CBCT. Practitioners should take a few steps to minimize scatter. Patients should be positioned according to the manufacturer’s instructions. For a CBCT of the jaws, this typically means having the occlusal plane parallel to the ground. Scatter tends to be worse if this recommendation is not followed. If the teeth do not need to be in occlusion during the scan, or if only one jaw is being imaged, it is good practice to have the patient bite lightly on a radiolucent material to separate the jaws. The authors have found that rope wax (Heraeus Kulzer, South Bend, IN, USA) works well, but other materials, such as Blu-Mousse (Parkell, Edgewood, NY, USA), work well too. The latter can also be used to take a bite registration if an accurate jaw relationship is necessary, and will serve the dual purpose of also separating the jaws. The purpose of separating the jaws is to reduce or prevent scatter from the opposing jaw extending to the jaw of interest, or at least making it easier to reduce or eliminate this scatter when performing a three-dimensional (3D) reconstruction.
The practitioner should keep in mind that the highest resolution is not always best. Besides increasing patient radiation dose, a higher resolution results in a much larger file size, which can make manipulation of the image much more difficult. Computer limitations aside, some applications can handle only up to a certain file size. Unfortunately, some practitioners are not conversant enough with their machines to feel comfortable adjusting the parameters to fit a particular study. They simply leave the scan parameters as the company representative sets them. This setting is often at the highest resolution and largest FOV, which result in an increase in radiation dose and scatter, and which may result in a file size that makes the image more difficult to manipulate.
This article discusses the use of CBCT. The authors would be remiss, however, in not disabusing the reader of a common misconception. CBCT does not allow one to do anything that cannot be done with conventional or medical CT. The reason that CBCT has in many respects replaced conventional CT is because of the lower radiation dose delivered by CBCT. Many dental and dental-related studies conducted using CBCT simply would not have been justifiable and previously were not performed using conventional CT, especially on children.
Clinical scenario 1: impacted teeth
One use of CBCT is to examine impacted teeth of the normal dentition, most commonly maxillary canines and mandibular third molars (M3), and supernumerary teeth, most commonly a mesiodens. In the case of maxillary canines, knowledge of the precise location of the impacted tooth, its relationship to the surrounding teeth, and the amount of buccal and palatal bone is critical for the orthodontist to determine whether space is available in which to rotate or translate the tooth into the arch and whether the procedure runs the risk of impinging on other teeth or extruding the impacted tooth outside the bone. The information may also be important to surgeons seeking to uncover the tooth with the intention of attaching a button and power chain, particularly when the tooth cannot be located clinically through observation or by palpation. Even while preparing this article, one of the authors received the following e-mail:
“Dr Friedland, I saw a 13-year-old female patient yesterday afternoon for canine exposure #11. After clinical examination and periapical radiographs… even using the so-called SLOB [buccal object] rule, it did not appear to me that the tooth in question moved either way… potentially indicating that it lay directly over the apices of the central and lateral incisor. Nonetheless a conservative palatal flap was raised… and small osseous resection undertaken… but to no avail. The decision was made to abort the procedure in an effort to ensure no harm was done to the dentition and given I could not discern the canine’s location—I simply closed up and explained to patient and mom. Well, I was hoping to have the patient take a CT scan to aid in locating the canine.”
Although an uncommon complication of mandibular third molar removal, with an incidence of 0.5% to 8% , mandibular nerve injury is nevertheless serious. One way to reduce the risk is for patients to undergo a CBCT. However, it is not necessary, nor practical from either a cost or radiation exposure perspective, to order a CT scan for every patient contemplating M3 removal. The development of selection criteria for ordering a CBCT scan must be further refined. Currently, a CT scan is appropriate when one or more panoramic signs suggesting an increased risk for inferior alveolar nerve damage are present .
Although CT or CBCT scans show the relationship between the canal and M3, the usefulness of these and other scans can be greatly enhanced for surgeons and patients by using software that creates a virtual 3D image of the teeth and jaws. Fig. 1 shows a virtual 3D image in which the bone, abutment teeth, metal framework, and teeth are in separate colors, known as masks . When using these proprietary softwares, separating the teeth from the bone in the 3D reconstruction is a simple task. Because teeth and bone have different amounts of mineralized tissue, with teeth being more highly mineralized, they produce radiographic images of different densities. In medical CT and CBCT, the density is measured in Hounsfield units. When selecting a 3D reconstruction above a certain Hounsfield unit, one can exclude bone from the image. Conversely, when excluding anything below a certain Hounsfield unit, one can exclude teeth from the image. Metal can be similarly easily separated. Clinicians can “hide” a mask to remove from the picture an object that may be obscuring anatomy that they wish to see more clearly in the 3D image. The image can also be rotated in any direction.
Because teeth have similar Hounsfield units, individual teeth cannot be separated on this basis. Rather, the technician must trace each tooth by going through the CT slices one by one. Therefore, an element of human interpretation is clearly involved in the reconstructions. Nevertheless, with an experienced technician the reconstructions are extremely accurate. Although a basic 3D reconstruction is easy to do, most oral surgeons will almost certainly find that it is not worth the time and trouble to do their own, especially reconstructions in which masking is desired. Furthermore, although a staff member can be trained to do basic and even advanced reconstructions, surgeons would be left in the lurch if that staff member left the practice. Therefore, the reconstructions are best referred out to third-party companies that specialize in this service. Most of these companies can accommodate images from virtually all CBCT scanners and are able to do the reconstructions in most all of the applications on the market.
Case Presentation 1
A 13-year-old woman presented to the orthodontist with an impacted maxillary left canine. Critical to treatment was whether the tooth could be brought into the arch. This determination could not be made from either the clinical examination or intraoral radiographs ( Fig. 2 ). A CBCT scan was taken ( Fig. 3 ) and a 3D reconstruction was performed ( Fig. 4 ). Based on the information, the orthodontist was able to determine that sufficient space was not present to bring the tooth into the arch. The 3D reconstruction was also useful to educate the patient and her parents about the reason for the determination. A mutual decision was made to remove the impacted canine. The information provided by the CT scan not only provided definitive information but also saved the patient treatment time in that it prevented an attempt to try to save the canine, which would have proved futile and might even have damaged adjacent teeth.
Case Presentation 2
A 26-year-old woman presented with a complaint of discomfort in teeth 30 and 31. Tooth 31 was known to be impacted and impinging on tooth 30. Periapical and panoramic images showed that the distal root of tooth 31 was dilacerated and seemed to be resting on the canal, which was difficult to identify with any reasonable degree of clinical certainty in the area of the mesial root ( Fig. 5 ). A CBCT was taken ( Fig. 6 ). The scan showed an especially intimate relationship between the mesial roots and the canal ( Fig. 6 D, cross-section number 67). The 3D reconstruction showed the nerve to be passing between two mesial roots of tooth 31 ( Fig. 7 ). The reconstruction also showed the relationship between the nerve and the roots with a clarity that could not be appreciated from the scan alone. One might have noticed on the periapical image that tooth 31 has two mesial roots.
Case Presentation 3
A 54-year-old woman was experiencing mild symptoms suggestive of a pericoronitis related to the mandibular right M3. A panoramic radiograph ( Fig. 8 A) suggested a close association between the mandibular canal and the roots of the tooth, and the oral surgeon ordered a CBCT scan (see Fig. 8 B, C). The 3D reconstruction showed the presence of two buccal roots and a lingual root, with the mandibular nerve passing between them. Based on the information, the surgeon and the patient jointly decided to not remove the tooth and to treat her symptoms conservatively. This case is another illustration of the usefulness of the additional information provided by the 3D reconstruction. Without the 3D reconstruction, the anatomy was difficult for the surgeon to fully appreciate, and it would have been extremely difficult for the surgeon to explain the anatomic configuration to the patient and to emphasize how it could have complicated removal of the tooth.
Case Presentation 4
An orthodontist referred a 9-year-old boy who had an impacted mesiodens. In addition to wanting to precisely localize the tooth, the orthodontist wanted to rule out fusion between the mesiodens and the right central incisor. The impacted mesiodens was visible in the axial view in Fig. 9 A (red arrow), whereas the cross-sections in Fig. 9 B showed it to be in intimate contact with the central incisor. A magnified view (see Fig. 9 C) showed both teeth to be fully formed, with each crown having its own enamel visible separately. The 3D reconstructions ( Fig. 10 ) showed each tooth to be fully formed, with the central incisor having normal anatomy. Based on these observations, the surgeon determined that the teeth were not fused and that the mesiodens could be safely removed. This removal was accomplished with no complications. Fusion of the roots would have dictated an entirely different course of treatment.
Although the profession must refine the selection criteria that would determine when a CT scan is indicated in cases of impacted molars and other teeth, the authors believe that a 3D reconstruction is an almost indispensable aide once the decision is made to obtain a scan. It provides additional information to the surgeon and helps obtain informed consent. The considerations for acquiring a 3D reconstruction are different from those that contribute to the initial decision of whether to obtain a scan. Because a scan has already been taken, a 3D reconstruction does not entail any additional radiation to the patient. Furthermore, the incremental cost of a 3D reconstruction can be justified in terms of the enhanced information that it adds, as illustrated in the cases described earlier.
Clinical scenario 1: impacted teeth
One use of CBCT is to examine impacted teeth of the normal dentition, most commonly maxillary canines and mandibular third molars (M3), and supernumerary teeth, most commonly a mesiodens. In the case of maxillary canines, knowledge of the precise location of the impacted tooth, its relationship to the surrounding teeth, and the amount of buccal and palatal bone is critical for the orthodontist to determine whether space is available in which to rotate or translate the tooth into the arch and whether the procedure runs the risk of impinging on other teeth or extruding the impacted tooth outside the bone. The information may also be important to surgeons seeking to uncover the tooth with the intention of attaching a button and power chain, particularly when the tooth cannot be located clinically through observation or by palpation. Even while preparing this article, one of the authors received the following e-mail:
“Dr Friedland, I saw a 13-year-old female patient yesterday afternoon for canine exposure #11. After clinical examination and periapical radiographs… even using the so-called SLOB [buccal object] rule, it did not appear to me that the tooth in question moved either way… potentially indicating that it lay directly over the apices of the central and lateral incisor. Nonetheless a conservative palatal flap was raised… and small osseous resection undertaken… but to no avail. The decision was made to abort the procedure in an effort to ensure no harm was done to the dentition and given I could not discern the canine’s location—I simply closed up and explained to patient and mom. Well, I was hoping to have the patient take a CT scan to aid in locating the canine.”
Although an uncommon complication of mandibular third molar removal, with an incidence of 0.5% to 8% , mandibular nerve injury is nevertheless serious. One way to reduce the risk is for patients to undergo a CBCT. However, it is not necessary, nor practical from either a cost or radiation exposure perspective, to order a CT scan for every patient contemplating M3 removal. The development of selection criteria for ordering a CBCT scan must be further refined. Currently, a CT scan is appropriate when one or more panoramic signs suggesting an increased risk for inferior alveolar nerve damage are present .
Although CT or CBCT scans show the relationship between the canal and M3, the usefulness of these and other scans can be greatly enhanced for surgeons and patients by using software that creates a virtual 3D image of the teeth and jaws. Fig. 1 shows a virtual 3D image in which the bone, abutment teeth, metal framework, and teeth are in separate colors, known as masks . When using these proprietary softwares, separating the teeth from the bone in the 3D reconstruction is a simple task. Because teeth and bone have different amounts of mineralized tissue, with teeth being more highly mineralized, they produce radiographic images of different densities. In medical CT and CBCT, the density is measured in Hounsfield units. When selecting a 3D reconstruction above a certain Hounsfield unit, one can exclude bone from the image. Conversely, when excluding anything below a certain Hounsfield unit, one can exclude teeth from the image. Metal can be similarly easily separated. Clinicians can “hide” a mask to remove from the picture an object that may be obscuring anatomy that they wish to see more clearly in the 3D image. The image can also be rotated in any direction.
