Every root canal is decided before the first bur touches the tooth. Whether a case is mine to treat, a case to refer, a tooth worth saving, or a tooth that should come out, all of that is case selection. And case selection is only as good as the information we bring to it.
For most of my career, that information came from a two-dimensional shadow of a three-dimensional object. In 2011 I installed my first CBCT scanner. About a month later, I stopped treating any tooth without a pre-operative scan. I have practiced that way ever since.
I like to describe a pre-operative CBCT as an open book exam of root canal anatomy. Every tooth we treat is a test of skill, experience, and applied knowledge. The scan is an individualized reference for that specific patient, one I can return to at any point during treatment when a question comes up that a 2D image can't answer. The more prepared we are going in, the better our odds coming out.
I want to be clear up front: routine pre-operative CBCT on every case is my personal practice philosophy, not the current Standard of Care. But whether you scan every tooth or only some, I believe 3D imaging changes case selection more than any other diagnostic tool we have. Here's why.
The Problem With Reading a 3D Tooth from a 2D Shadow
A periapical radiograph is a single beam passing through a three-dimensional object. That comes with built-in limitations no amount of experience can fully overcome.
Geometric distortion. Over-compensate vertical angulation and the tooth foreshortens. Under-compensate and it elongates.
Superimposition. Roots overlap each other. The zygomatic arch, tori, the incisive canal, and the mental foramen can hide pathology, or mimic it.
Interpretation is less objective than we'd like. Goldman and colleagues found only about 50% agreement between examiners on what counted as apical pathosis on a 2D image, and only 75-83% agreement with their own earlier reads months later. Decades later, Tewary et al. (J Endod 2011) found only fair agreement among residents, endodontists, and an oral radiologist reading the same 150 digital periapicals.
I've seen what superimposition does in practice. A mental foramen sitting right at the apex of a premolar can look exactly like a periapical lesion on a periapical film. The incisive canal can do the same thing to a maxillary central. One sagittal or coronal CBCT slice settles the question.
The flip side matters just as much. CBCT has been shown to identify 54% more periapical lesions than 2D imaging (Lofthag-Hansen et al., 2007), with 61% greater accuracy (Stavropoulos et al., 2007) and 62% greater sensitivity (Estrela et al., J Endod 2008).
The number that should get every clinician's attention comes from a study right here in Chicago. Ee, Fayad, and Johnson (J Endod 2014) had three board-certified endodontists review 30 cases with periapical radiographs, then review the same cases with CBCT two weeks later. The treatment plan changed in roughly 62% of cases. If more than half of our plans change once we see the tooth in three dimensions, that says a lot about how much we were missing in two.
Five Ways 3D Imaging Changes Case Selection
You see the anatomy before you commit to it
Root canal anatomy has been studied since Hess published his work in 1917, and published averages by tooth type are useful. But averages describe populations, not the patient in your chair. In my experience, every tooth has genuinely complex morphology. You simply can't judge a tooth's true complexity from a 2D image. A scan shows you what you're signing up for: an extra distolingual root on a mandibular molar, a canal that splits and rejoins in a lower incisor, a C-shaped root, a dens invaginatus, severe curvature in a plane the periapical never captured. It also shows you how much of the pulp chamber is left, and how deep you'll have to go to reach it. Even on a tooth that looks completely "normal" on film, the scan adds value. Knowing whether two canals merge before the apex changes how I approach and interpret patency chairside. I'd much rather know that before I start than discover it with a file.
You confirm the source before you treat it
The right treatment on the wrong tooth is still the wrong treatment. CBCT is especially valuable when symptoms are poorly localized, when clinical findings contradict the radiograph, or when a previously treated tooth hurts but shows nothing on a periapical. It can also trace a sinus tract. A chronic infection breaks through the cortical plate, and that low-density path can be followed through the volume back to the source tooth. In my view, that makes gutta-percha tracing and the SLOB rule largely obsolete. A coronal slice tells you not only buccal versus lingual, but the exact vertical level of what you're looking at.
You know the prognosis before you start
Case selection isn't just "can I treat this?" It's "should anyone?" Some of the most important findings on a scan are the ones that tell you a tooth isn't a good candidate at all. A vertical defect with complete loss of supporting bone around a root points to either a root fracture or a true combined periodontal-endodontic lesion. Telling them apart may not always be possible, and it may not matter, because both carry an unfavorable prognosis. A deep, isolated probing depth on the buccal or lingual may not show at all on a 2D image. CBCT can also reveal a missing buccal plate or a through-and-through lesion. That information changes the conversation with the patient. It's the difference between offering a root canal with confidence and honestly discussing extraction, or a periodontal consult, before anything irreversible happens.
Retreatment becomes a targeted decision
Non-surgical retreatment is one of the clearest indications for a pre-operative scan. When a previously treated tooth fails, the question is why. CBCT often answers it: a missed canal, an untreated root, or damage from a prior access you'd never see on film. I published a case in the Journal of Endodontics that illustrates this. A patient presented with a painful, previously treated lower second molar under a cast post and core, supporting a partial denture. The periapical showed a radiolucency near the distal root. The CBCT showed something more specific: an untreated distolingual root (radix entomolaris) with a lesion isolated to that root, while the mesial apices looked healthy. Instead of a full disassembly, we offered selective root retreatment, treating only the root with disease. The scan let us plan a precise access through the existing core to that one canal. Without 3D imaging, that option doesn't exist. The scan also shows you hidden risk. I've opened referred cases where a prior access had gouged deep into a wall, leaving it dangerously thin. Seeing that on the scan before I pick up a bur tells me where to be careful, and whether the tooth is restorable at all.
You know when to refer
For general dentists, this may be the biggest advantage of all. Difficulty in endodontics doesn't add up, it compounds. A posterior tooth, a malpositioned tooth, limited opening, a pulp chamber masked by a restoration, and a deep target depth each add difficulty on their own. Combine two or more and the difficulty climbs exponentially. A scan makes those variables visible before the case starts, when nothing irreversible has been done. That's exactly the right moment to decide a case falls outside your comfort zone. Referring before access is good dentistry. Referring after a perforation is damage control.
CBCT Limitations
I'm an advocate for CBCT, but it isn't magic, and it isn't a replacement for anything.
It supplements 2D imaging; it doesn't replace it. Right now, the rubber dam and 2D radiographs are the accepted Standard of Care equipment in endodontics. Skip the 2D images and you could technically fall below that standard, no matter how much more the scan shows. Good 2D fundamentals still matter, and the supplies behind them, from film holders and XCP positioning devices to barrier sleeves, are easy to source through suppliers like Net32.
Artifact is real. Posts, cores, and metallic restorations cause beam hardening and aliasing that can make part of a scan non-diagnostic at that level. You have to recognize it when you read the volume.
You own the whole volume. The clinician who orders a scan is responsible for interpreting all of it, not just the tooth in question, even findings outside endodontics. When something is beyond you, refer it promptly to an oral and maxillofacial radiologist. But a radiology over-read won't make your treatment decision for you. It may flag gross pathology without identifying the specific etiology you need.
The dose is higher. CBCT delivers more radiation than conventional dental imaging. I justify it by the mid-treatment images it saves and the irreversible errors it helps avoid. Use the smallest field of view that answers the question; for most single-tooth complaints, a limited field of view is all you need.
Coverage is still catching up. CDT codes exist for CBCT, but most insurers don't yet cover it. That cost falls on the patient or the practice, and it's a real barrier to adoption.
The guidelines are more conservative than I am. The AAE/AAOMR joint position statement does not support routine CBCT for screening without clinical signs and symptoms. My argument is that its own first indication, potential complex morphology, applies to essentially every tooth. Reasonable clinicians disagree, and you should read the statement yourself.
If you don't have a scanner, a third-party imaging center is a perfectly good option. And if you decide a scan is necessary for a case but the patient declines, my personal view is that it's worth discussing a different treatment option altogether. There's no reason to add to our profession's limitations by proceeding without information that's available at minimal risk.
We've Been Here Before
New technology in dentistry tends to move through three phases. First comes disbelief. Then the "OK, it works, but I don't need it" phase, where clinicians accept that the technology does what it claims but feel their results are already good enough. Finally comes mass acceptance, when we look back and wonder how we ever did it any other way.
The operating microscope is the perfect example. It arrived in endodontics in 1982 and took roughly 15 years to become mandatory in specialty programs. Early adopters recommended it only for the "hard" cases. Then Buhrley and colleagues (J Endod 2002) showed clinicians located canals 71% of the time with the microscope, 62% with loupes, and 17% with no magnification. Today nobody questions using a microscope on every tooth.
The argument for the microscope was simple: we see more, so we treat better. CBCT shows us far more than the microscope ever could, and it shows us before we start. Selective use of CBCT has already reached mass acceptance. Routine pre-operative use, in my estimation, is somewhere in the late "OK, it works" phase.
Ours is the first generation of clinicians with routine access to individualized 3D anatomy before treatment begins. How we use it will shape where endodontics goes next. My philosophy is simple: understanding anatomy preoperatively is better than reacting to anatomy after the fact. Good case selection starts there.
Dr. William Nudera, DDS, MS, is a Diplomate of the American Board of Endodontics and the author of NuEndo: ReThinking Endodontics. He practices at Root Canal Care Center in Highland Park, IL, and Specialized Endodontic Solutions in Bloomingdale, IL, and co-hosts the For the Love of Endo podcast.
References
Goldman M, Pearson AH, Darzenta N. Endodontic success—who's reading the radiograph? Oral Surg Oral Med Oral Pathol. 1972;33(3):432-437. https://doi.org/10.1016/0030-4220(72)90473-2
Goldman M, Pearson AH, Darzenta N. Reliability of radiographic interpretations. Oral Surg Oral Med Oral Pathol. 1974;38(2):287-293. https://doi.org/10.1016/0030-4220(74)90070-x
Tewary S, Luzzo J, Hartwell G. Endodontic radiography: who is reading the digital radiograph? J Endod. 2011;37(7):919-921. https://doi.org/10.1016/j.joen.2011.02.027
Lofthag-Hansen S, Huumonen S, Gröndahl K, Gröndahl HG. Limited cone-beam CT and intraoral radiography for the diagnosis of periapical pathology. Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 2007;103(1):114-119. https://doi.org/10.1016/j.tripleo.2006.01.001
Stavropoulos A, Wenzel A. Accuracy of cone beam dental CT, intraoral digital and conventional film radiography for the detection of periapical lesions. An ex vivo study in pig jaws. Clin Oral Investig. 2007;11(1):101-106. https://doi.org/10.1007/s00784-006-0078-8
Estrela C, Bueno MR, Leles CR, Azevedo B, Azevedo JR. Accuracy of cone beam computed tomography and panoramic and periapical radiography for detection of apical periodontitis. J Endod. 2008;34(3):273-279. https://doi.org/10.1016/j.joen.2007.11.023
Ee J, Fayad MI, Johnson BR. Comparison of endodontic diagnosis and treatment planning decisions using cone-beam volumetric tomography versus periapical radiography. J Endod. 2014;40(7):910-916. https://pubmed.ncbi.nlm.nih.gov/24935534/
Hess W. Formation of root-canals in human teeth (Thomas NG, Redlich H, trans.). J Natl Dent Assoc. 1921;8(9):704-734. Originally published in German, 1917. https://doi.org/10.14219/jada.archive.1921.0179
Nudera WJ. Selective root retreatment: a novel approach. J Endod. 2015;41(8):1382-1388. https://doi.org/10.1016/j.joen.2015.02.035
Sousa Melo SL, Fayad MI, Gohel A, et al. AAE and AAOMR joint position statement: use of cone-beam computed tomography in endodontics 2025 update. Oral Surg Oral Med Oral Pathol Oral Radiol. 2026;141(1):126-135. https://doi.org/10.1016/j.oooo.2025.09.013
Buhrley LJ, Barrows MJ, BeGole EA, Wenckus CS. Effect of magnification on locating the MB2 canal in maxillary molars. J Endod. 2002;28(4):324-327. https://doi.org/10.1097/00004770-200204000-00016
