In a 2008 article on cone beam volumetric tomography (CBVT) and dentoalveolar applications, Tyndall and Rathore wrote: ‘It is in the area of endodontic applications that the literature has proved most fruitful to date.’1 This statement is even truer today than in 2008. A review of the literature has demonstrated that, in many cases, CBVT is more efficacious than traditional forms of 2-D imaging. Endodontic applications of CBVT include the diagnosis of periapical lesions due to pulpal inflammation, identification and localization of internal and external resorption, the detection of vertical root fractures, the visualization of accessory canals, and elucidation of the causes of non-healing endodontically treated teeth. Prior to 2008 most published articles on CBVT applications in endodontics were either case reports or in vitro studies. Since that time more well designed clinically related scholarly activity has been published. This article attempts to survey the field of CBVT applications in endodontics and provide the readers with an overview of what has been found. The authors hope that this knowledge will form a foundation for appropriate clinical decision making with specific reference to selection criteria for the endodontic applications of CBVT.
The basis for this growing evidence of the efficacy of CBVT in endodontic applications is found in the classic studies on the limitations of 2-D radiography for the detection of periapical lesions by Bender and Seltzer.2,3 Their studies revealed that in order for a lesion to be visible radiographically, the cortical plate of bone must be involved. These findings, revealing the difficulty of detecting periapical lesions, have been consistently verified in subsequent studies since that time. A review by Huumonen and Orstavik summarized much of that research, postulating that such limitations exist, partly because of the 2-D nature of intraoral radiographs where clinical or biologic features may not be reflected in radiographic changes.4 While there have been many advances in receptor and X-ray tube technologies since the first dental radiograph was taken in 1896, there have been essentially no changes in imaging geometry for the dentition since that time. Even panoramic imaging is still a form of 2-D imaging and has not contributed significantly to endodontic applications of X-ray imaging. CBVT is a relatively new type of imaging geometry that more adequately describes and illuminates the 3-D anatomy of the teeth and jaws. It is no surprise that such technology has resulted in a near revolution in imaging for endodontically related dental problems.
As the review below proceeds, and the case examples are shown, the reader should be aware of the paucity of literature based on double blind clinical trials using more robust, in vivo research methodologies. Since these types of time consuming studies generally use technologies that are out of date upon publication, scholars and clinicians must base case management and selection criteria decisions on the somewhat lower level of studies extant today.5 Case examples of some of the applications of CBVT are illustrated in Figs. 1–8.
Current CBVT systems and endodontic applications
In 1972, Sir Godfrey Hounsfield announced an invention that used image reconstruction developed in the 1960s by Alan Cormack. This new invention eventually became known as computed tomography and it transformed medicine as well as diagnostic radiology such that three-dimensional imaging is now the standard of care for trauma and pathology in the medical field. In 1998, Mozzo et al.6 introduced a new volumetric CT machine using cone beam technology useful for maxillofacial imaging. The need for three-dimensional accuracy in pre-implant planning combined with a desire to decrease the radiation dosages from conventional CT were the reasons for continuing changes in what has come to be known as cone beam volumetric tomography.
The technology of CBVT is described elsewhere in this supplement and will not be reviewed here. What follows is a review of CBVT examples currently on the market with potential for endodontic applications.
Large field of view (FOV) units are from 15 cm to 23 cm and are most useful in the assessment of maxillofacial trauma, orthodontic diagnosis and treatment planning, TMJ analysis and pathologies of the jaws. The NewTom 3G, 5G, the iCAT next generation and the Kodak 9500 are such examples of units used for craniofacial imaging. These machines may also provide smaller FOV options. Medium FOV encompasses those CBVTs with a FOV of 10–15 cm which are useful for mandibulo-maxillary imaging and are used primarily for pre-implant planning and pathological conditions. Machines in this category include the Galileos by Sirona, Gendex CB-500 the NewTom VGi, 3D Accuitomo 170 and the My-Ray Skyview. Small FOV units, aka limited FOVs, are becoming increasingly popular and encompass FOVs less than 10 cm with some as small as 4 x 4 cm in size. These units are appropriate for dentoalveolar imaging and are most desirable for endodontic applications. Examples of CBVTs with small FOVs include but are not limited to the Kodak 9000 3D, the Sirona Orthophos XG 3D, the Veraviewepocs 3D and Accuitomo from Morita as well as the Prexion. Many of the CBVTs listed above are available in multiple FOVs and voxel sizes.
Radiation dosages have received extensive media coverage lately and are a very real concern for patients. Published values of effective dose can give a broad indication of the level of detriment to health from radiation exposure. In describing the radiation risks attributed to CBVT, it can be helpful to compare effective dose to radiographic exams that are common in dentistry. Ludlow et al. used the 2007 ICRP weightings and found a direct digital panoramic radiograph to be 14.2 μSv while a full mouth series of radiographs (FMX) with F-speed film and rectangular collimation to be 34.9 μSv.7 One way to help patients further understand the doses that they are receiving is to equate dental radiographic examinations to the amount of background radiation that one receives naturally on a daily basis. According to the United Nations Scientific Committee on the Effects of Atomic Radiation, the average worldwide background radiation is about 2.4 mSv (2400 μSv) per year or approximately 6.7 μSv a day. Therefore, a panoramic radiograph would equate to just over two days of background radiation while the FMX described above would be equivalent to 5.2 days of background radiation. CBVT dosages vary considerably based on the FOV, the exposure beam type (pulsed vs. continuous), technique settings (mAs, kVp), beam geometry and the amount of basis projections. The literature also varies depending on whether the 1990 or the 2007 ICRP weighting factors are used. Table 1 gives the effective doses of several small volume CBVTs using the 2007 ICRP weighting factors broken down by panoramic and daily per capita background radiation doses. The Somatom 64 multidetector CT (MDCT) used in medicine is provided as a comparison. Although there is a reduction in dose, it is important to follow the principles of ALARA (As Low As Reasonably Achievable). The overall diagnostic benefit to the patient must outweigh the radiation risks of receiving the exam.
Table 1. Imaging and dosimetry characteristics of selected small field of view CBVT units
|Limited FOV CBVTs available in the United States as of June 2011||Voxels in mm||Field of View||Exposure||AEffective Dose in μSv||BDigital Panoramic Equivalent||CNo. of days of annual per capita background radiation|
|3D Accuitomo FPD 170||0.08, 0.125, 0.160, 0.250||4 × 4 cm||Continuous||243 ||3.1||6.4|
|Kodak 9000 (C) 3D||0.076||5 × 3.7 cm||Pulsed||*5.3–38.3 depending on region||0.38–2.7||0.79–5.7|
|Promax 3D||0.1, 0.2||8 × 8 cm||Continuous||228–122 depending on settings||2–8.7||4.17–18.2|
|PaX-Uni 3D (OS)||0.12 × 0.2||5 × 5cm||Not Published||244||3.14||6.5|
|Veraviewepocs 3D||0.125–0.2||4 × 4 cm, 4 × 8 cm||Continuous||3sFOV = 30–40 depending on FOV||2.1–5.2||4.5–6.0|
|PreXion 3D (Standard exposure)||0.2||3.2 inches diameter (8.1 × 7.5cm)||Continuous||*189||13.5||28.2|
|Scanora 3D||0.13–0.35||6 × 10 cm scan diameters||Pulsed||*76||5.4||11.3|
|Comparison with Somatom Sensation32 row/64 slice MultiDetector CT||0.6||Body width × 12 cm||Continuous||*860||61.4||128.3|
|Comparison with Somatom 32 row/64 slice MultiDetector CT w/CARE dose 4 D||0.6||Body width × 12 cm||Continuous||*534||38.1||80|
Several recent investigations have demonstrated the accuracy of CBVT and are briefly summarized below. CBVT allows for an accurate three-dimensional representation of the scanned area. Geometric accuracy has been proven since the introduction of the CBVT.6 Kobayashi et al.12 compared limited volume CBCT to spiral CT in measuring mandibular ‘lesions’ made in cadaver mandibles. Their data showed that limited volume CBVT could measure distances accurately. These findings agreed with a study by Lascala et al.13 that analysed the accuracy of linear measurements obtained by CBVT to those of digital calipers in eight dry skulls. They found that the measurements between anatomical sites of the facial area taken with CBVT were statistically similar to actual measurements. They concluded that measurements could reliably be made with CBVT. An in vivo study further validated the accuracy of linear measurements as well as volumetric measurements in CBVT by conducting two consecutive experiments with defects of known sizes. Pinsky et al.14 first used a cast acrylic block with holes of various sizes and then used a human mandible with 21 engineered simulated defects. They found that the mean linear accuracy was smaller than 0.1 mm in the acrylic block and less than 0.3 mm in the mandible. Using a voxel size of 0.2 mm, they observed that the overall measurements were either less than or equal to two voxels. They further concluded that CBVT errors are small and not clinically significant. One study investigated the accuracy of CBVT and intraoral digital radiographs for the detection of bony and infrabony defects. The study found that CBVT had an overall more accurate assessment than digital intraoral radiographs in detecting both types of defects.15
Summary of current literature addressing CBVT and endodontics
There are many reports in the literature of the benefits of CBVT, particularly in endodontics. Endodontic applications include localization and detection of broken instruments, non-healing root canals needing retreatment, root resorption, root fractures, understanding canal morphology, trauma, detection of periapical lesions and the extent of extruded root canal material. The technology has been widely accepted and is now being used for research and clinical purposes.
Patients with endodontic problems can pose a serious challenge in terms of diagnosis and treatment planning. The exact problem is often hard to discern when a patient may have symptoms without any radiographic signs of further periapical disease. It is important to correctly identify the problem and plan accordingly for reasons discussed above.2,3 There are multiple limitations to two-dimensional radiographs, such as superimposition of three-dimensional anatomy as well as possible exposure or geometric errors.16 Tyndall et al.1 found CBVT superior for almost all endodontically related uses when compared to conventional 2-D radiographic surveys. A study by Sanfelice and colleagues17 used CBVT instead of histological sectioning to compare four different instruments used to flare the cervical third of a root. A clinical study by Cotton et al.18 provided case examples of various applications of a high resolution limited CBVT in endodontics. It proved the usefulness of three-dimensional imaging in detecting a missed canal in a non-healing root canal, identification of root fractures, pathological conditions that were not of endodontic origin, the extent, type and prognosis for root resorption lesions, as well as the assessment of anatomy in close proximity to root apices. A case report by Tsurumachi and Honda19 used CBVT to help in the detection, localization and surgical pre-planning of a broken instrument. They felt that while periapical films give good detail mesiodistally, they are inadequate to give detail in the buccolingual dimension. Therefore, it was concluded that CBVT helped not only in detecting the exact position of the instrument, but also led to a safer surgical approach. Limitations of viewing structures was also noted by Low et al.20 who felt that the maxillary molars in particular were difficult to assess with 2-D films. When comparing the diagnosis of periapical lesions, anatomical relationships and pre-planning for apical surgeries with CBVT versus periapical radiography, it was discovered that CBVT revealed 34% more lesions than periapical radiographs. They were also able to appreciate expansion of the lesions into the maxillary sinuses, thickening of the sinus mucosa, missed canals as well as apicomarginal communications much easier with CBVT. This finding is similar to the study conducted by Lofthag-Hansen et al.21 which found 38% more apical lesions on CBVT than with two periapical radiographs taken at 10 degree horizontal angles. This study also found sinus membrane thickening more often with CBVT than with periapical films. In fact, CBVT revealed additional relevant information in 32 of the 46 cases involved. The impact of 3-D imaging was evaluated using CT in the diagnosis and treatment planning of non-healing root canals.4 It was discovered that of the 39 teeth observed, 30 had a second mesiobuccal (MB2) canal present and 27 of these MB2 canals had been missed and remained unfilled; 22 of the 27 teeth with missed canals had periapical lesions. The authors felt that knowledge of the size and extent of periapical lesions, buccal and lingual cortices as well as the maxillary sinus boundaries were important when deciding on a surgical approach. This potentially significant information could be provided by CBVT.
Root fractures are quite difficult to detect on 2-D radiographs unless the X-ray beam passes directly along the fracture line.22 The clinician must rely on a set of symptoms that cast suspicion on a diagnosis of a fractured tooth. It becomes a challenge to confidently recommend a course of action when one is not completely sure of the exact diagnosis. A systematic review conducted by Tsesis and coworkers23 evaluated articles on vertical root fractures from 1971 to January 2010 in order to further characterize their appearance on radiographs. The most frequent radiographic feature noted in the various articles was a combination of periapical/perilateral radiolucencies that they referred to as a halo sign. However, they concluded that evidence based data on the clinical and radiographic signs leading to a diagnosis of vertical root fracture (VRF) was lacking. CBVT research addressing the problem of horizontal and vertical root fractures continue to be carried out as clinicians search for a better way to diagnose these confusing entities. Bornstein et al.24 observed 44 permanent teeth in 38 patients that sustained trauma resulting in horizontal root fractured teeth. It compared periapical and occlusal films with limited volume CBVT to evaluate the location and angulation of the fracture line. The study found that horizontal root fractures could be easily seen in all 44 teeth with the CBVT. A case study by Orhan and colleagues25 discussed an instance where CBVT was able to determine whether root resorption was involved with a horizontally fractured front tooth. They used 3-D imaging and found that no periradicular pathosis or resorption was present. CBVT was instrumental in diagnosing the tooth as a spontaneously healed root fracture and the patient was able to retain his tooth without further treatment. Research by Hassan and Metska et al.26 focused on the comparison of VRF detection on CBVTs and periapical radiographs. They were specifically assessing the effect of root canal material on the ability of the modalities to detect these types of fractures. They found the overall accuracy of CBVT scans to be superior to periapical radiographs. However, they did note that the detection of VRFs was limited by the contrast to noise ratio as well as the voxel size which was 0.25 mm in their study. It was also discovered that the presence of root canal material did not affect the overall accuracy of CBVTs but it did reduce its specificity. The authors postulated that the beam hardening or streak artefacts observed with root canal materials may have made the observers less confident in diagnosing the VRFs. Various thicknesses of VRFs were evaluated in a study by Ozer et al.22 that used a limited volume CBVT and a voxel size of 0.125 mm to observe fractures down to 0.2 mm. It was concluded that CBCT was statistically superior to digital radiography for all thicknesses of VRFs noted in the study.
Three-dimensional imaging has many additional benefits including characterization of lesions for pathological purposes. CBVT is a potentially useful tool in the identification of margins for surgical biopsies as well as in differential diagnoses. However, a CBVT is unable to give a clinician a definitive diagnosis the way that a histological biopsy can. Concerns over previous papers that suggested that CBVTs can be used instead of histopathology to differentiate radicular cysts from granulomas led Rosenberg et al.27 to publish a study to evaluate the truth behind these claims. This study included 45 patients and had two radiologists and two pathologists independently examine the samples. It observed the consistency of the radiology reports and found a weak inter-rater reliability (κ = 0.14) while the inter-rater reliability of the pathologists was quite strong (κ = 0.79). The study compared the two radiologists’ findings with the gold standard and found that their accuracy was 51% for the first radiologist and 61% for the second. Therefore, it was concluded that histopathology is still the gold standard for differentiating a radicular cyst from a granuloma. CBVT has improved many areas of endodontics but for this particular diagnostic task a biopsy is still necessary.
Identification of root canals and root canal morphology is another area where CBVT has been shown to be superior to 2-D imaging. A paper by Weine et al.28 reported the prevalence of MB2 canals in maxillary first molars. They found that 51.5% of maxillary first molars exhibit some type of MB2 canal. They further explain that it is often difficult to detect these canals ahead of time with intraoral radiographs and can be considered a possible cause in unexplained failure of treatment. Degerness et al.29 studied the dimensions, anatomy and morphology of the mesiobuccal root canal system in maxillary molars by sectioning and describing 150 teeth. Their study resulted in reporting a higher incidence of canals in the mesiobuccal root of the first maxillary molars than in the previous study. They found that 20% of their sample had one canal, 79.8% had two canals and 1.1% had three canals. They concluded that a thorough understanding of the complicated root canal system in maxillary molars would improve endodontic therapy. Kottoor and colleagues30 presented a case report of unusual anatomy in the maxillary first molar with seven root canals that were diagnosed with surgical operating microscopes and confirmed with CBVT. The authors felt that the unusual anatomy was proven with 3-D imaging and led to the successful case management. The accuracy of CBVT and other modalities in identifying root canal morphology have been compared to the modified canal staining and clearing technique by Neelakanton and coworkers.31 They analysed 95 teeth to identify the number of canals found with each method. CBVT was able to correctly identify the canals 99.71% of the time. The inter-rater agreement between CBVT and the modified canal staining and clearing technique was 99% for the five observers (three endodontists and two radiologists) and only 82% for digital radiographs. In fact, observers missed two or more canals with digital radiographs in 23.8% of teeth. Further validation of CBVT as a tool in exploring the root canal anatomy was observed in the study by Michetti et al.32 This study compared CBVT with histological sections viewed under an optical microscope. They found on average a strong to very strong correlation between the CBVT and the histological sections (r area = 0.928 and r diameter = 0.890). The authors concluded that CBVT was a reliable and non-invasive way to view the root canal anatomy.
The presence of periapical lesions can determine the treatment outcome of a tooth. CBVT has proven to be beneficial in diagnosing periapical lesions that intraoral periapical radiographs failed to show. A study by Lofthag-Hansen et al.21 helped to establish the accuracy of and increased visualization using CBVT over periapical radiographs. This study found that 62% more apical lesions were located with CBVT than with intraoral periapical radiographs. It further showed that lesions with a mean mesial distal width of 2.8 mm and a mean buccolingual dimension of 4.4 mm were not detected on periapical radiographs but were noticed on CBVT. A study by Sjögren et al.33 found that the success rate for teeth that have vital or non-vital pulps but no periapical lesion is 96%. However, cases that have a necrotic pulp and a periapical lesion have a success rate of 86% and that drops to 62% for root-filled teeth with periapical radiolucencies. The absence of periapical lesions on intraoral radiographs does not mean that the apices are free of lesions. CBVT has been shown to diagnose these lesions better because there is not a superimposition of cortical bone over the lesion.34 Stavropoulos and Wenzel35 studied the accuracy of CBVT, digital intraoral and conventional films in detecting periapical lesions in pig jaws. They were able to detect artificially created bone defects statistically more often with CBVT than with the other two modalities. They postulated that the low sensitivity of the intraoral modalities was due to the fact that the artificially created defects were limited to the cancellous bone. This agrees with previous studies.2,3 Nakata et al.36 presented a case report of a patient with poorly localized pain in the right maxillary molar region. Panoramic and intraoral radiographs were unable to determine the cause of the patient’s pain. A small volume CBVT was able to reveal a 4 x 4 mm lesion on the distobuccal root of a previously root canal treated maxillary first molar. The author further explained the necessity of knowing the correct pathological conditions, anatomical structures and positional relationships in order to give the best quality of endodontic treatment. Paula-Silva and colleagues37 evaluated periapical lesions with periapical radiographs and CBVT and compared them to histopathological findings as the gold standard. They found that apical periodontitis was discovered in 71% of roots with periapical radiographs, 84% with CBVT and 93% with histology. They found an overall accuracy of 92% with CBVT when considering sensitivity, specificity, positive predictive values and negative predictive values. It is possible that root-filled teeth once thought to have healed by intraoral radiographic standards in fact still have periapical lesions when viewed with CBVT. This technology may change our view of what constitutes a ‘healed’ tooth and the length of time in which teeth are evaluated post treatment.34,38
The detection and management of internal and external root resorption can be a challenging task and one in which CBVT is well suited. The knowledge gained from three-dimensional imaging can help in diagnosing the size of the defect as well as its proximity to the root canals and ultimately the prognosis of the tooth. Two studies by Patel et al.39,40 pay particular attention to the use of CBVT for this purpose. In 2007, the author discussed two cases where CBVT helped with diagnosing the true extent and management of external cervical resorption. It was noted that angled periapical radiographs using the parallax technique can be helpful in trying to determine the location of a lesion, as well as whether it is internal root resorption or external root resorption, but is unable to help determine the depth or extent of such lesions. On radiographs, internal root resorption is noted as a smooth, well defined radiolucency that may be spindle shaped and contiguous with the root canal. The fact that the lesion does not change positions when viewed on two angled radiographs is also characteristic of internal root resorption. External root resorption will not appear to be as well defined and the unaltered outline of the root canal may be observed through the defect. These lesions will appear to change positions when viewed on the two angled periapical radiographs. The author felt that CBVT not only showed the full extent of the lesions but also allowed the clinician to be more confident in their treatment approach, as well as give them a more realistic prognosis for each tooth in question. Two years later their second study was reported and evaluated both internal and external root resorption cases and compared them to a control group examining the ability of CBVT to accurately detect lesions. More importantly, it was the first study to examine the impact that CBVT made in determining the correct treatment plan for the patient. CBVT was found to have perfect accuracy in detecting and diagnosing the different resorptive lesions when compared to periapical radiographs. The observer’s ability to choose the correct management of the lesion was 60% for intraoral radiographs and 80% for CBVT when compared to a consensus committee. The study had a small sample size but dealt with a difficult diagnostic task. Clinical studies of this nature must also depend on a silver standard in order to preserve the tooth in question. However, this report marked the beginning for studies that wish to move beyond the question of accuracy and try to determine the actual benefit of this technology to the patient. Scarfe et al.16 has stated that ‘the absence of prospective randomized clinical trials underlines the need for further research on the treatment outcomes related to CBVT applications in endodontic practice’.
The American Association of Endodontics and the American Academy of Oral and Maxillofacial Radiology have recently released a joint position paper41 discussing the use of CBVT in endodontics. The findings of both organizations are reasonably applied across the globe and will be summarized herein.
It was suggested that small FOV units are better suited to endodontics because their inherent small voxel sizes result in higher resolution images (down to 0.076 mm) and less radiation dosages than the larger FOV options. An important consideration is patient selection criteria. CBVTs should not be used for screening purposes and not every patient needs a 3-D image. Cases should be chosen on an individual basis depending on the patient’s history, clinical examination and inability to obtain adequate diagnostic information from 2-D images. As stated previously, it is important that the diagnostic benefit to the patient exceed the risk of radiation. CBVT should be limited to difficult endodontic cases such as:
Identification of accessory canals, complex morphology, root canal system anomalies including determination of root curvature, such as in the case of maxillary molars.
Cases of contradictory or non-specific signs and symptoms.
Poorly localized symptoms associated with a previously treated tooth.
Anatomic superimposition unresolved with 2-D imaging.
Diagnosis of non-endodontic pathology.
Assessment of intra or postoperative complications.
Diagnosis of dentoalveolar trauma.
Localization of root resorption.
Pre-surgical planning for apical surgeries as well as for dental implants.