Reading Cone-Beam CT Scans
What are we missing that we should not be missing?
Christos Angelopoulos, DDS, Msc
Cone-beam computed tomography (CBCT) was introduced to dentistry approximately 18 years ago and has been almost instantly embraced by the dental profession. For the first time, oral health professionals had an opportunity to assess an area of concern with multiple thin sections and from all possible views with small, compact, in-office devices, eliminating the disturbing noise caused by overlapping surrounding structures as occurred with traditional projectional dental imaging modalities (periapical and panoramic imaging). It also significantly enhanced diagnostic efficiency. Admittedly, CBCT was not our first experience with tomographic (sectional) imaging, which was available in dentistry since the late 70s with computed tomography (CT). This technology has improved tremendously over the last 30 years with fast image acquisition devices, capable of acquiring several sections simultaneously as with multi-detector CT (MDCT). However, medical CT was not regarded as a modality close to dentistry for routine dental diagnostic applications; these devices were not easily accessible, the radiation to the patient was sometimes excessive, and the image detail was often not up to our expectations. These reasons, along with a reasonable cost and thus the radiologic examinations, explain the spread of CBCT in dentistry.
Soon after the introduction of CBCT, and despite the excitement about it, it was realized that dentists were not that familiar with tomographic imaging. Apart from a few exceptions, like specialists in oral and maxillofacial radiology and surgery, who had advanced training in sectional imaging modalities, no other had any formal training or experience in such imaging, which admittedly is quite a challenge.1 The biggest concern was the fact that acquiring an imaging volume, as with CBCT, seems to be revealing more information about the maxillofacial region in comparison to the narrow zone of osseous and dental-mostly- structures seen in panoramic and intraoral imaging. Moreover, clinicians who are unfamiliar and poorly trained with these kinds of images sometimes handle this.
Various scientific organizations expressed these concerns with a number of publications over the last few years and underlined the value of mandatory education for the CBCT operator beyond that provided by the CBCT vendor, which is often the only training acquired.2
As addressed above, when an imaging volume is acquired for the investigation of a specific diagnostic concern (eg, possible pathology in the right mandible), unavoidably, the imaging data may include a wider zone of structures based on the field of view (FOV) selected by the operator of the CBCT. In our example, this may include the entire mandible, the floor of the mouth, the suprahyoid neck, and even a small portion of the maxilla. The inclusion of additional anatomical areas may reveal pathological entities that are either silent or have not progressed yet in order to provoke clinical findings; these are known as occult or incidental entities.
Incidental findings are not uncommon and sometimes seen in other dental diagnostic images like panoramic and intraoral radiographs; however, they are identified mostly in the osseous structures because these are the only ones shown in these images. Moreover, dentists are quite familiar with osseous pathology and these are usually promptly addressed.
On the contrary, in CBCT imaging volumes the reported prevalence of incidental findings is rather high.3 This is certainly associated with its volumetric nature and, also the selected FOV. Expanded FOV CBCT examinations may include anatomical areas like the maxillary and other paranasal sinuses, the nasal cavity, the nasopharynx, portions of the skull base and neck, etc. Occult entities may appear in these areas even though the CBCT scan was acquired for an unrelated reason. Often, these entities are readily identifiable and may range from anatomical variants and developmental anomalies to more significant pathological entities, which may need to be addressed sooner rather than later. Sometimes the incidental entities are inconclusive and may require either further investigation or periodic evaluation to assess possible progression. Naturally, the main concern about incidental findings is possible professional, ethical, and even legal issues that may be involved in their recognition but also their future management. This has not yet been adequately addressed. For example, is it important if the clinician who acquired the CBCT scan did not diagnose a mild sinus inflammation or a nasal septum deviation? What if the undiagnosed entity was carotid artery atheromatosis? If these incidental findings were identified in both cases, should the clinician be taking the same actions? If yes, what are those? If in both cases the clinician informs the patient about the findings and refers the patient to a specialist for further assessment and possible treatment, is this the right thing to do? Is it the right thing to do informing a patient about a finding that is considered of low significance like a developmental anomaly or an anatomical irregularity that appeared as incidental entities in a CBCT scan or is it over-diagnosis? The answers to the above posed questions are unclear mostly because of a lack of guidelines.
Prevalence of Incidental Disease
The prevalence rates of the CBCT incidental findings reported in the literature vary and are dependent upon the age of the patients examined, the variability of the FOV selected, and the different categories of incidental entities recorded by different researchers. Albeit direct comparison of the results of these studies may not be meaningful because of the different types of findings, the respective authors might consider incidental entities of importance; the total prevalence exceeded 50% in most of the studies.4
Categorizing them by anatomical location (ie, paranasal sinuses, nasal cavities, skull base, soft tissue of the neck, cervical spine, etc.) and by significance (ie, the need for additional actions like periodic assessment or a referral) may be a systematic approach to record incidental findings.
Below, we report the prevalence of incidental findings in a series of 561 CBCT scans acquired over a period of 18 months. All scans were made with the same CBCT scanner (GALILEOS, Dentsply Sirona, www.sirona.com) and a fixed FOV (15 cm x 15 cm) for routine dental assessment or dental implant placement. All scans were reviewed by a maxillofacial radiologist and a written report was generated and maintained in the patients’ dental records. The incidental entities found were categorized by means of significance based on the need for periodic evaluation or to refer the patient to a different clinician to address the incidental pathological entity. The most frequently met incidental entity in our review was a deviated nasal septum (83.1% of the cases reviewed). This is considered an abnormal condition or developmental anomaly in which the nasal septum is not centrally located in the nasal cavity but is shifted towards one side resulting in uneven nasal chambers. In most cases this is mild and requires no action. In some cases this is rather severe and may result in significant narrowing of the nasal passageways and thus difficulty in breathing (Figure 1).
Thickening of the mucosal lining of the maxillary sinuses (and other paranasal sinuses) is also a common finding (49.4% of the cases) of low significance that requires no action.
Tonsillar calcifications or tonsilloliths (13.7% of our cases) are mineral deposits accumulated in the tonsils that often appear as clusters of calcifications superficially located in the tonsillar tissues (Figure 2). Most frequently, they are seen in the palatine tonsils, bilaterally, at the level of the floor of the mouth. They are rather benign and require no action unless they are large, which is rare. Other incidental findings of low or moderate significance may be seen in Table 1.
Narrowing of the airway is the most common incidental entity that may require some action to be taken. This is most frequently associated with a retruded tongue, soft palate, or even tonsillar hypertrophy, and may be one of the causes of sleep apnea (Figure 3). Pending the severity of the airway restriction and the resulting apnea, a referral to an ENT physician or a sleep disorder specialist may be the appropriate management.
Another occult pathological entity, the identification of which is often incidental, is sinus inflammation (11.4% of our cases showed extensive sinusitis). The maxillary and other paranasal sinuses may show different degrees of concentration of inflammatory tissues in CBCT scans, which may be asymptomatic or the related symptoms are low grade (Figure 4). These symptoms may progress and significant morbidity may be established (sinus congestion, coughing, pressure headaches, etc.) if the affected paranasal sinus is completely occupied by inflammatory tissues. If such a pathological entity is recognized, an ENT referral is required.
One of the most significant incidental pathological findings that may be seen in a CBCT scan of the maxillofacial region or the mandible is the extracranial carotid artery calcifications resulting from atheromatosis (Figure 5). Their most common location is the carotid bifurcation, where the common carotid artery splits into the external and internal carotid arteries. This occurs at the level of C3 and C4 vertebral bodies, posterior-laterally to the airway, an anatomical region that is included in FOV of the CBCT scan. Carotid atheromatosis is a silent disease that is associated with a high risk of stroke in older individuals and if identified by means of presence of calcifications in the respective blood vessels, the patient and the patient’s physician need to be informed for further investigation because CBCT is not the appropriate diagnostic tool to quantify the associated vascular obstruction. A detailed listing of the various incidental findings and their prevalence is seen in Table 1 and Table 2.
Managing Incidental Pathological Entities Encountered in a CBCT Scan
This is a tricky issue, primarily because there are no clear guidelines. The reasonable approach would be to inform the patient about the finding and outline clearly what the next steps might be, such as further investigation if the status of the incidental entity is unclear or a referral to a different specialist for treatment if this is considered necessary. This seems proper if the incidental finding is considered significant, but what if it is not? Wouldn’t it be quite stressful for the patient to be informed or guided for further management for some tonsillar calcifications or a deviated septum, which bears no impact on his or her breathing function? The answer is “yes.” In such a case do we even need to inform the patient? There are physicians who support emphatically that unnecessary or excessive medical information may be causing psychological distress to the patient and that diagnosticians should refrain from discussing incidental findings with the patient if the disclosure bears little or no benefit to the patient; they claim that this is the way to prevent over-diagnosis,5 not to mention the extra cost involved if additional diagnostic studies are required.
On the contrary, others support that it is our duty to inform the patients about all incidental findings as it is their right to know because there is no clear and legitimate definition of what constitutes an incidental finding. Consequently, it is the patient’s right to decide on possible exposure to any risks that an incidental entity may entail.6
Final Thoughts
The concerned dental professional should consider a few key guidelines in regard to occult disease and CBCT image diagnosis in general. Follow a system and review thoroughly the imaging volume in a standardized manner, irrespective of the reason that the CBCT scan was ordered. Do the same for all CBCT scans. List all findings that may be suspicious of pathology or some kind of abnormal appearance of known anatomical areas or tissues. This includes incidental disease as well. A prerequisite for this is becoming familiar with the appearance of disease and anatomical variants. This is based mostly on experience with CBCT and training.
Generate a standardized report of all your findings based on anatomical location after you narrow down the ones that clearly are of low significance. If a reported abnormality seems to be an anatomical variant, report it as such. This may include possible recommendations for management or the best possible future action. When in doubt, consult with a specialist in oral and maxillofacial radiology for a second opinion and guidance. In such a case, share the entire volumetric data set rather than selective images. The oral and maxillofacial radiologist may provide not only diagnostic advice but also may act as a link to other diagnostic imaging specialists.
In the end, the best way to deal with incidental disease comes back to our professional judgment, responsibility, and knowledge.
Disclosure
The author has no relevant financial relationships to disclose.
References
1. Angelopoulos C. Anatomy of the maxillofacial region in the three planes of section. Dent Clin North Am. 2014;58(3):497-521.
2. Carter L, Farman AG, Geist J, et al. American Academy of Oral and Maxillofacial Radiology executive opinion statement on performing and interpreting diagnostic cone beam computed tomography. Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 2008;106(4):561-562.
3. Dobele I, Kise L, Apse P, et al. Radiographic assessment of findings in the maxillary sinus using cone-beam computed-tomography. Stomatologija. 2013;15(4):119-122.
4. Miles DA. Clinical experience with cone beam volumetric imaging report of findings in 381 cases. Computerized Tomography. 2005; 416-424.
5. Volk ML, Ubel PA. Better off not knowing: improving clinical care by limiting physician access to unsolicited diagnostic information. Arch Intern Med. 2011;171(6):487-488.
6. Brown SD. Professional norms regarding how radiologists handle incidental findings. J Am Coll Radiol. 2013;10(4):253-257.
