It Looked Like a Cyst. It Wasn't. The Imaging Was the First Warning Sign Nobody Read.

Every oral surgeon has read a referral that ends with the same three words: "for cyst enucleation." A well-corticated radiolucency around an impacted third molar, a periapical lesion that has been quietly enlarging for years, a swelling that comes and goes — and somewhere along the way, before the CBCT is even opened, a working diagnosis gets written down. Most of the time, it holds up. Sometimes it doesn't. 

The gap between "looks like a cyst" and "is a cyst" is where real complications begin — a second surgery, a delayed referral, or a diagnosis that arrives months later than it should have. We looked back at three cases that came through as routine cystic lesions and turned out to be an odontogenic keratocyst, an ameloblastoma, and a malignancy. In every one, the correct diagnosis was already sitting in the CBCT dataset from the very first scan. Nobody hid it. Nobody read it either. 

Why the First Read So Often Gets It Wrong 

A cyst is common. An ameloblastoma is not, and a primary malignancy of the jaw is rarer still. In a busy practice, that arithmetic quietly shapes how every new lesion gets looked at. When nine similar-looking radiolucencies out of ten turn out to be exactly what the referral says, the tenth one tends to get read through the same lens as the other nine — not because anyone is careless, but because that is simply how pattern recognition behaves under time pressure and repetition. 

CBCT does not fix this by itself. A three-dimensional dataset only adds diagnostic value if someone looks past the referral diagnosis and examines the image on its own terms — asking what the bone is actually doing, rather than confirming what the clinical note already suspects. The scan can hold every feature needed for an accurate diagnosis and still be read in a way that misses all of them, simply because the reader already knew what they were expected to find. 

The three cases below cover the width of that spectrum: a benign but locally destructive cyst, a benign tumour with a real recurrence rate if mismanaged, and a malignancy that had already outgrown its "infected cyst" label by the time imaging was ordered. In each, the features that mattered were visible in the original scan. They were not faint. They were only easy to miss because a diagnosis had already been written on the referral slip before anyone looked. These are composite descriptions drawn from patterns that turn up repeatedly across referred CBCT scans, not identifying details of any one patient, but every finding described below is one we see often enough to treat as a recurring lesson rather than a rare anomaly. 

Case One: The Cyst That Grew the Wrong Way 

A man in his early thirties was referred for enucleation of what the referring Dentist had noted as a "dentigerous cyst" around an impacted mandibular third molar, picked up incidentally on a panoramic radiograph taken for orthodontic assessment. There was no pain, no paraesthesia, and no history of infection — the kind of quiet presentation that rarely raises alarm on its own. 

On CBCT, the lesion extended well beyond the crown of the impacted tooth, running forward along the body of the mandible rather than pushing outward into the surrounding soft tissue. The buccal cortex over this stretch was thinned and, in one small segment, frankly perforated — yet the jaw itself had barely changed shape. A cyst that had been present long enough to erode through cortex would ordinarily be expected to show some outward bowing of that same cortex; here, the bone had been hollowed out from within far more than it had been pushed aside. On the axial slices, the internal content was not uniformly low-density the way straightforward cystic fluid usually appears — there were patchy areas of slightly higher attenuation scattered through the lesion. 

The tooth itself sat in a reasonably favourable position relative to the lesion, and the differential sitting quietly next to "dentigerous cyst" on the Radiologist's report was odontogenic keratocyst. Histopathology after enucleation confirmed it. The distinction mattered for one very practical reason: OKCs carry a materially higher recurrence rate than dentigerous cysts when treated with enucleation alone, and the surgical and follow-up plan for the two are not meant to be the same. 

Case Two: The Ameloblastoma Filed Under "Just a Cyst" 

A woman in her early forties presented with a slowly enlarging, painless swelling of the posterior mandible that she had first noticed roughly a year earlier. Her Dentist's referral read "dentigerous cyst, for enucleation," based on an orthopantomogram showing a well-defined multilocular radiolucency associated with an unerupted second molar. 

The CBCT told a more complicated story. Expansion was present on both the buccal and lingual cortical plates — not just one side, which is the more common pattern with a passively expanding cyst. In places the buccal plate had thinned to little more than an eggshell, and a small area near the inferior border showed early perforation. Internally, the lesion was genuinely compartmentalized rather than a single smooth cavity with a few incidental septa; coronal sections showed several distinct loculations of varying size, giving the coarse, soap-bubble appearance that gets mentioned so often in oral pathology texts but is easy to underweight on a first pass. Two adjacent molar roots showed resorption that was angular and tapered, rather than the smoothly rounded blunting more typical of slow, benign pressure resorption. 

Given the bilateral cortical expansion, the compartmentalized internal architecture, and that particular root resorption pattern together, ameloblastoma sat well above dentigerous cyst on the differential before the patient ever reached the operating room. Resection with a margin of clinically normal bone followed, rather than simple enucleation — and histopathology confirmed a solid and multicystic ameloblastoma. Enucleation alone, on the strength of the original referral diagnosis, would very likely have left disease behind. 

Case Three: When "Infected Cyst" Was the Wrong Read Entirely 

The third case was the one that should have raised concern earliest and didn't. A man in his late fifties came in with pain, mild facial swelling, and some looseness of two adjacent lower molars. His general Dentist's working diagnosis was an infected periapical cyst secondary to a non-vital tooth; a course of antibiotics brought partial relief, and imaging was ordered only when the pain returned within weeks. 

The CBCT findings did not fit that story once looked at closely. The lesion's margins were irregular and non-corticated across most of its extent — a sharp contrast to the smooth, well-defined border almost always seen with an inflammatory cyst that has had time to organise a bony response around itself. The periodontal ligament space around both mobile teeth was uniformly widened along the length of the root, not just at the apex, which is not the pattern a simple periapical process produces. Most tellingly, the buccal cortex in the area was neither thinned nor expanded the way it had been in the previous two cases — it was disappearing outright, with patchy areas of frank cortical destruction and no compensatory bone remodelling at the margins. 

That combination — an infiltrative, non-corticated margin; diffuse periodontal ligament widening around more than one tooth; and cortical loss without expansion — does not belong to a benign cystic process, however painful or however textbook the clinical story sounds. An incisional biopsy was arranged without further delay, and it confirmed a primary intraosseous carcinoma of the mandible. The antibiotic course, in hindsight, had bought the disease three additional weeks. 

The Three Cases, Side by Side 

Set next to each other, the pattern is easier to see than it was for any single case in isolation.

Case	Working Diagnosis on Referral	Confirmed Diagnosis	Cortex & Margin on CBCT	Root / PDL Finding
Case 1	Dentigerous cyst	Odontogenic keratocyst	Thinned, focally perforated cortex; minimal outward expansion	Adjacent tooth largely unaffected
Case 2	Dentigerous cyst	Ameloblastoma (solid/multicystic)	Buccal + lingual expansion, cortex thinned to perforation	Angular, tapered root resorption
Case 3	Infected periapical cyst	Primary intraosseous carcinoma	Irregular, non-corticated margin; cortical destruction, no expansion	Diffuse PDL widening around two teeth

A table like this flattens a lot of clinical nuance, and none of these features stand alone as diagnostic on a first pass through a busy reporting list. What it does make clear is that all three lesions were telling the reader something different from the referral note attached to them — and that information was available on day one, not discovered in retrospect once histopathology came back. 

The Common Thread: What Bone Does Under Pressure Versus Under Attack 

Placed side by side, the three cases trace a single underlying pattern. A benign cyst grows slowly enough that bone has time to remodel around it — the cortex is pushed outward and thinned, but stays intact and corticated for a long time before it finally fails. A benign tumour like ameloblastoma grows faster and more aggressively but is still, in a sense, negotiating with the bone around it — cortex expands in more than one direction, thins further, and eventually perforates, but expansion is still the dominant story throughout. A malignancy simply outpaces the bone's ability to respond at all. There is no time for expansion, no corticated border gets the chance to organise, and the periodontal ligament — the one structure that should reliably widen only where a tooth is actually inflamed — starts widening in places it has no clinical business doing so. 

None of this is meant as a checklist to memorise from a blog post. The same features genuinely overlap between entities in ways that require real training to interpret with confidence, and CBCT has honest limits — it does not reliably show true soft-tissue extent, and staging a suspected malignancy still calls for cross-sectional CT or MRI, not a CBCT field of view designed for dentoalveolar detail. What CBCT does show, reliably, is that the bone is telling a different story from the one written on the referral form — and that story is available on the very first scan, not only in retrospect once someone finally goes looking for it. A radiologist who has read enough of these cases develops a feel for that difference almost intuitively; a clinician reading their own scan occasionally, alongside a full day of chairside work, understandably does not have the same repetition to draw on, and that gap is exactly what a dedicated second read is meant to close. 

What This Means Before the Scalpel Ever Touches Bone 

For an Oral Surgeon or OMF Specialist, the CBCT is usually already sitting in the file by the time a surgical plan gets discussed with the patient. That timing matters. A Radiologist's report that simply confirms the referring diagnosis is not adding much to the case; a report that has actually been read with pathology in mind — independent of what the referral slip says — is doing the job a formally reviewed and reported scan is supposed to do. 

This is also where general Dentists play a genuine role, well before a case ever reaches a surgeon's chair. A lesion that looks routine enough to plan for extraction and curettage in-house is worth a second look for exactly that reason — the cases that go wrong are, almost by definition, the ones nobody thought to double-check. We have written before about the wider category of findings that turn up on CBCT scans ordered for something else entirely, and the same principle holds here too: an unexpected finding deserves a systematic read, not a quick confirmation of what was already expected. 

The uncomfortable truth in all three cases here is that nothing about the imaging process failed. The scans were taken correctly, at appropriate resolution, with an adequate field of view. What changed the outcome was simply someone taking the time to read the bone itself rather than the referral note attached to it. That is a workflow question as much as a clinical one — it depends on whether the person interpreting the scan has both the training and the time to look past the expected answer, case after case, without the referral diagnosis quietly setting the terms of the review. 

The Referral Diagnosis Is a Starting Point, Not a Verdict 

None of this is a criticism of referring Dentists — a working diagnosis on a referral slip is meant to communicate context, not to close the differential. The problem starts when that same working diagnosis quietly becomes the lens through which the imaging gets read, by anyone downstream, including the Radiologist. We touched on this broader gap between recognizing a lesion and truly interpreting it in an earlier piece on oral pathology and CBCT, and it is worth restating here: recognizing that something that looks like a cyst is the easy part. Working out what it actually is — and what its cortical, marginal, and periodontal behaviour is trying to tell you — is a different skill entirely, one that a handful of case descriptions in a blog post can only gesture toward, not teach in full. 

We have also seen, repeatedly, that the questions a Dentist is able to ask after opening a CBCT scan — rather than simply reading the one-line summary — say a lot about how confidently a practice can catch cases like these before they turn into surgical surprises. 

Reading the Bone, Not Just the Referral 

At DMD Imaging, every CBCT scan we receive — regardless of what the referral diagnosis says — is formally reviewed and reported by a qualified Dental Radiologist before it goes back to the treating clinician. That is not a marketing line; it is the only mechanism that catches cases like the three above at the first scan instead of the second surgery. An MDS-qualified Dental Radiologist reading with the pathology in mind, rather than confirming a working diagnosis, is the difference between "looks like a cyst" and knowing, with reasonable confidence, whether it actually is one. 

We've written elsewhere about why that distinction between a Radiologist and a Radiographer matters for the report you actually receive, and these three cases are as good an illustration of that gap as any we've come across. 

For Oral Surgeons and OMF Specialists who want to build this pattern recognition into their own reading of CBCT data — rather than relying solely on the report that comes back — our CBCT Surgery & Pathology course works through the systematic interpretation framework that these three cases can only hint at here: how margin quality, cortical behaviour, internal architecture, and root or periodontal ligament changes fit together into a coherent differential, case after case, rather than as isolated observations noticed too late.

Frequently Asked Questions

• Can a CBCT scan misdiagnose a jaw cyst?

  A CBCT scan itself does not diagnose anything — it captures data. Misdiagnosis happens at the interpretation stage, when a lesion's imaging features are read to confirm an existing clinical impression rather than assessed independently. In each of the cases above, the scan contained the correct answer from day one; the initial read simply didn't go looking for it.

• What CBCT features suggest a jaw lesion is not a simple cyst?

  Features worth a closer look include cortical expansion on more than one side of the jaw, thinning or perforation of the cortex out of proportion to the lesion's apparent duration, an irregular or poorly corticated margin, unusual patterns of root resorption, and periodontal ligament widening that extends beyond a single tooth's apex. None of these findings are diagnostic alone, but together they shift the differential away from a straightforward benign cyst.

• How is an ameloblastoma different from a dentigerous cyst on CBCT?

  Both can present a well-defined radiolucency around an unerupted tooth, which is exactly why they get confused. Ameloblastoma more often shows expansion of both the buccal and lingual cortical plates, a coarse multilocular internal pattern, and root resorption with a sharper, more angular profile, whereas a dentigerous cyst typically expands in one direction and produces smoother, more rounded resorption where roots are involved.

• Can oral cancer be mistaken for a dental cyst on imaging?

  Yes, particularly in its earlier stages, when pain and swelling can mimic an infected cyst and antibiotics may produce partial, temporary relief. The imaging clue is usually the margin and the bone response: malignant lesions tend to show irregular, non-corticated borders and cortical destruction without the expansion typically seen in benign cystic disease, along with periodontal ligament changes around more than one tooth rather than a single apex.

• Should every jaw lesion on CBCT get a second Radiologist opinion?

  Not every lesion needs a second opinion, but any lesion where the clinical picture and the imaging don't fully agree — or where the working diagnosis was formed before the CBCT was reviewed independently — is worth one. A report from a qualified Dental Radiologist who reads the scan without anchoring to the referral diagnosis is the safeguard that catches cases like these before they reach the operating room.

• Does CBCT alone confirm whether a jaw lesion is malignant?

  No. CBCT can raise or lower suspicion based on margin quality, cortical behaviour, and periodontal changes, but a confirmed diagnosis still requires histopathology, and true staging of a suspected malignancy relies on cross-sectional imaging such as CT or MRI rather than a CBCT field of view built for dentoalveolar detail. CBCT's role is to flag which lesions need that next step urgently, not to replace it.