My tooth hurts but nothing shows on the X-ray: could it be a cracked tooth or a root fracture?

A standard X-ray records how much the beam weakens as it passes through the tooth. A crack, though, is usually a hairline gap; unless the beam passes along the line of that gap, the density of the rest of the tooth masks it and no trace is left on the film. That is why what an X-ray often shows is not the crack itself but its consequences: bone loss around the root, or the shadow of inflammation at the root tip. In the cracked tooth registry study run by the US National Dental Practice-Based Research Network with 209 dentists, which followed 2,858 patients for three years, only 2% of the cracks visible to the eye could also be made out on an X-ray. The words 'there is nothing on the X-ray' do not mean there is no crack; they mean the X-ray cannot answer this question.

The same study also shook a long-held belief. The first symptom that comes to mind with a cracked tooth is a sharp pain the moment you release your bite, and a review published in the Japanese Dental Science Review also describes this as the typical feature of a cracked tooth. In the registry study, however, 37% of the teeth were painful with cold and 16% on biting, and in 13% the pain came on by itself. The authors describe this as the study's most surprising finding. Clenching or grinding, the tooth being a molar, and the crack being on the side of the tooth that faces the back of the mouth were found to be linked with symptoms.

The classification most reviews use comes from the American Association of Endodontists and divides lengthwise tooth cracks into five groups. The first is craze lines, hairline marks confined to the enamel; they do not cause pain. The second is a fractured cusp: a crack that starts at the chewing surface, runs into the dentine and ends at the neck of the tooth, separating off one cusp. The third, the cracked tooth in the strict sense, is an incomplete fracture running from the chewing surface towards the root that has not yet split the tooth in two. The clinical review describes this group as both the most common and the one with the most variable course. The fourth is a split tooth, where the crack has divided the tooth into two pieces from end to end. The fifth is a vertical root fracture: a fracture running lengthwise in the root. In the first three groups the tooth can often be kept; in the last two, the conversation often shifts from saving the tooth to what will replace it.

No single test makes the diagnosis; the dentist puts several clues together. In a bite test, the cusps are loaded one at a time to find the point that hurts on release, and the review notes that this test carries a risk of extending the crack. A cold test shows the state of the nerve. When a strong light is shone through the tooth, a crack blocks the light. Dye shows where the crack is but not how deep it goes. The same review states that cracks narrower than 18 micrometres are hard to see with the naked eye, which is why a microscope is used. In the registry study, the main findings that led dentists to treat were decay, pain on biting, the crack being visible on an X-ray, and pain that came on by itself.

Because a cone beam CT (CBCT) scan shows the tooth in three dimensions, it looks at first like the tool that will solve the problem, but its limits are clear. The Japanese review writes that even with suitable settings, cracks finer than 50 micrometres are hard to make out on a CBCT scan, and that the bright streaks created by fillings make the job harder still. With vertical root fractures the picture is slightly better. A meta-analysis published in BMC Medical Imaging in 2024 finds a sensitivity of 0.51 for standard X-rays and 0.70 for CBCT; roughly speaking, CBCT picks up seven of every ten roots that really are fractured, and a standard X-ray five. In the same analysis, there is no significant difference between the two methods in their tendency to give false alarms, and 19 of the 23 studies in the review were carried out not in the mouth but on extracted teeth.

Things get harder still with a root-filled tooth: in a subgroup of the same BMC analysis, the sensitivity of CBCT is 0.76 in root-filled roots and 0.91 in roots without a filling. In a meta-analysis published in the International Journal of Dentistry in 2025, which pools 100 laboratory studies, a root filling on its own did not significantly change the result; the marked drop came from a metal post, which brought sensitivity down to 0.56, and incomplete fractures are also clearly harder to pick up than complete ones (0.52 against 0.75). Images taken with a smaller voxel size, meaning a higher resolution, improve the result in teeth with posts. The authors also state plainly that the image quality achieved in the laboratory cannot be matched in the mouth, and that well-designed studies in real patients are needed. What this means in practice: a clear CBCT scan does not rule out a fracture, and a line that looks like a fracture is not a verdict on its own.

Vertical root fractures are seen mostly in teeth that have had root canal treatment. The review in the Journal of Personalized Medicine writes that the reported frequency in root-filled teeth varies across a very wide range depending on the study, from 3.69% to 25%, and lists removing too much dentine, over-widening the canal when preparing the space for a post, and excessive pressure during root filling among the risk factors. The symptoms are quiet: severe pain is rare, and a dull discomfort is more usual. The clinical signature is a narrow, deep pocket at the single point that lines up with the fracture (in 64% to 93% of cases) and a draining opening that appears closer to the gum margin than to the root tip (in 35% to 42%). On an X-ray, a halo around the root or a shadow shaped like the letter J is typical, yet 13% to 14% of fractures leave no shadow on the X-ray at all. According to the review, the gold standard for a definite diagnosis is opening the area with a minor operation and seeing the fracture directly.

The reason a vertical root fracture so often ends with losing the tooth is not the size of the fracture but its position. The line runs inside the bone, along the surface of the root, and opens up a pathway for bacteria that cannot be sealed; as the fracture progresses, the bone around it resorbs too. Approaches such as bonding the fracture, or extracting the tooth, joining the pieces outside the mouth and putting it back, have been tried, but the evidence rests on small case series: in a series of 26 teeth treated with this last method, 88.5% of the teeth were still in the mouth at one year and 59.3% at five years. In molars with several roots, removing the fractured root can be an option. The review writes that the long-term results of these approaches are not yet known.

The other side of the picture matters too: most cracked teeth are not lost. In the registry study, 98% of the teeth were still in the mouth after three years, only 3% broke, and 91% of those breaks were partial fractures that did not involve the root, such as a lost cusp. For cracked teeth that need root canal treatment, two systematic reviews cited in the review in Clinical and Experimental Dental Research report survival of 84.1% to 88% over 12 to 60 months, and a pocket linked to the crack before treatment is one of the factors that worsens the outcome most.

The point where this topic meets our field is a narrow one. Diagnosing and treating a single cracked tooth is a job for your own dentist or an endodontist. The topic only concerns us if more than one tooth has been lost or is on the verge of being lost; the decision about the single tooth then becomes part of a plan that deals with many teeth together.

The distinction below shows the general line seen in the reviews and the registry study. For your own tooth, the decision is made by the dentist who examines it.

The sequence below shows the steps generally followed for a tooth whose X-ray is clear but whose symptoms carry on. Some of the steps are completed in the same appointment; a definite diagnosis is sometimes made only once a filling has been removed or the area has been opened.

1. 1

   The history of the pain

   Does the pain come when you bite down or when you let go? What does cold do? Has the tooth had root canal treatment or a post? Do you clench your teeth? These answers narrow down the possibilities a great deal.

2. 2

   Bite and cold tests

   The cusps are loaded one at a time to find the point that hurts on release. A cold test shows whether the nerve is alive. Do not try the bite test yourself at home on something hard; the force can extend the crack.

3. 3

   Light, dye and magnification

   Light shone through the tooth stops at a crack, dye makes the crack's position stand out, and a microscope shows lines the naked eye cannot pick out. None of them tells you on its own how deep the crack goes.

4. 4

   Pocket measurement

   The pocket is measured point by point around the tooth. A narrow pocket that suddenly deepens at a single point strengthens the suspicion of a root fracture. If there is a draining opening in the gum, a thin gutta-percha cone, the material used in root fillings, is placed into it and an X-ray is taken; if the cone runs parallel to the root surface, a fracture becomes a strong possibility.

5. 5

   X-rays and, if needed, a CBCT scan

   A standard X-ray shows inflammation at the root tip and the pattern of bone loss. A CBCT scan is more sensitive for root fractures, but a root filling and a metal post can cast shadows on the image.

6. 6

   Looking directly

   The definite answer often comes from removing the old filling and tracing the crack or, where a root fracture is suspected, from opening the area with a minor operation and looking at the root. In the registry study, at least one crack was also seen within the dentine in 89% of the teeth prepared for a restoration; the plan can change at this stage.

7. 7

   Questions to ask when discussing the result

   Which cusp is the crack in, and how far down does it go? How many millimetres deep is the pocket, and is it at a single point? Is the nerve alive? Is there any shadow from a filling or post on the CBCT scan? Which restoration is being recommended, and why? If extraction is needed, how does waiting affect the bone?

8. 8

   Loss that has spread beyond one tooth

   The decision about a single tooth is your own dentist's. If you also have other missing teeth or teeth that cannot be saved, and the conversation is turning to replacing more than one tooth, send the panoramic X-ray or CBCT scan you have through the form; the dentist who will carry out the treatment reads the image, and we will reply within 24 hours. This is a preliminary assessment; the final decision is made after an examination and a CBCT scan.

Which route is right depends on the type of crack and how far down it goes. All of the following are decisions for your own dentist or an endodontist.

With a cracked tooth, the risk cuts both ways: a fracture that goes unnoticed can progress, but not every crack needs treatment either. These are the traps people fall into most often.

The sequence below shows the general course for a tooth that has been diagnosed with a crack and is to be kept. Timings vary with the person and the procedure carried out.

1. The days around the diagnosis

   Do not chew anything hard on that side. Your dentist may fit a temporary crown or a band to protect the tooth.

2. The interim period

   Some dentists monitor the tooth with a temporary crown for a few weeks before the permanent restoration. If the pain settles during this time, that suggests the nerve can be kept; if it continues or gets worse, root canal treatment comes into the picture.

3. The first few months after the permanent restoration

   Brief sensitivity to cold can be expected. Pain with cold that lingers, or pain that comes on by itself or wakes you at night, can mean the nerve has been affected and needs checking.

4. Over the years

   In the registry study, most changes in the cracks were seen in the first two years. Regular check-ups, and a night guard if you clench your teeth, are the basis of this period.

You will not find figures here. With a cracked tooth, the total is set by what the diagnosis turns out to be and how many steps it takes to reach it:

The steps needed for a diagnosis

A picture that becomes clear with an examination and a standard X-ray is not the same job as one that needs a CBCT scan, a microscope or a minor exploratory operation.

The type of restoration

A filling, an onlay and a full crown need different amounts of time, materials and laboratory work. Asking which one is being recommended, and why, is the quickest way to understand a quote.

Whether root canal treatment is needed

If the nerve stays alive, the work ends with the restoration; if the nerve has been affected, root canal treatment followed by a crown is added.

The plan for the gap if the tooth is lost

If the tooth is extracted, what replaces it is a separate piece of planning. If the loss spreads to more than one tooth, that plan takes on a much wider scope than a single tooth.