A dark band at the neck of a tooth on a dental X-ray can look like decay, but sometimes it is an effect of how the X-ray passes through the tooth. One of the main effects of this kind is called cervical burnout. Other optical effects can make a healthy surface look shadowed too, which is why a 2026 review advises that X-ray findings are always checked against a careful clinical examination.
This article explains the common lookalikes, what the research says about how often they are misread, and how a dentist tells them apart from real decay.
Why a healthy tooth can show a dark band
How bright each part of an X-ray image looks depends on how much of the beam that part absorbs. Enamel absorbs a lot and looks bright. At the neck of the tooth, between the edge of the enamel and the top of the bone, the beam passes across the curved side of the tooth through less tooth structure. A 2026 systematic review describes the result: a localised area that looks darker and can mimic a cavity near the junction where the enamel ends and the root surface begins. The review notes the effect can be more obvious on teeth with more than one root, especially when the roots do not line up in the image (Dioguardi and colleagues, 2026).
Three effects that can look like decay
| Effect | Where it shows | Why it happens |
|---|---|---|
| Cervical burnout | The neck of the tooth, between the enamel edge and the bone | The X-ray beam passes through less tooth structure there, so the area looks darker |
| Mach band effect | Where enamel meets dentine, or where a filling meets tooth | An optical illusion: the eye exaggerates the contrast where a bright area sits next to a darker one |
| Triangular radiolucency | Mostly the side of an upper molar that faces the front of the mouth (the mesial surface) | A triangle-shaped dark area linked with the Mach band effect and the shape of upper molars, often mistaken for decay |
All three are described in the 2026 review. A literature survey from the University of Hong Kong found that dental papers usually mention the Mach band effect only in their discussion, mostly in relation to decay at the enamel and dentine junction and around fillings, and that very few had studied it directly (3 of 77 publications) (Yeung, 2022).
How often these effects are misread
The 2026 review could include only five studies, and rated the certainty of the evidence as low to very low. Its figures are about triangular radiolucencies and the Mach band effect, not cervical burnout. Within those limits, its findings were:
- Triangular radiolucencies that were not decay were seen on about 26% of upper molars across two studies (270 of 1,021 teeth, a simple count; the review's weighted estimate was about 31%).
- They were more common on baby molars. In German children aged 11 to 12, they appeared on 60% of upper second baby molar surfaces examined, compared with about 25% of upper first permanent molars.
- In three studies that tested how these two effects were read (one in children's teeth, one on extracted teeth and one using a simulated case), they led to a false diagnosis of decay or a fracture in about 13% of observations (60 of 464).
An older study of 25 full-mouth X-ray sets from a Saudi dental school found cervical burnout more often on upper teeth than lower, most often on upper incisors and, in the lower jaw, on canines. In premolars, the authors reported that 70% of the burnout seen on periapical X-rays (which show the whole tooth) was not visible on the bitewing view (Rahmatulla and Wyne, 1995). This suggests the type of image can change what is seen.

X-rays can also miss early decay
The opposite mistake happens too. A systematic review of 117 studies found that X-rays had low sensitivity for decay in general, including early decay, but moderate to high specificity. Sensitivity was higher for decay that had reached the dentine, and higher again for decay that had formed a cavity between the teeth (Schwendicke and colleagues, 2015). In plain terms, an X-ray is better at confirming more advanced decay and less reliable for the earliest stage, which is one reason it is read alongside a look at the tooth. Our article on what a dental X-ray shows that a visual exam cannot covers the other side of this.
How a dentist tells burnout from decay
healthdirect describes how decay is diagnosed: the dentist examines your teeth, asks about symptoms and may take an X-ray. For the lookalike effects, the 2026 review advises that X-ray findings should always be checked against a careful clinical examination. In practice that can involve:
- Looking closely at the surface in question, where it can be seen.
- Checking where the shadow sits. Burnout commonly appears between the enamel edge and the top of the bone.
- Comparing a different view of the same tooth. In the Rahmatulla study, much of the burnout seen on premolars in periapical X-rays was not visible on the bitewing view.
If the area is early decay rather than an effect, healthdirect notes that decay which has not made a hole can be treated with fluoride paste or varnish, together with brushing with fluoride toothpaste and changes to diet, and that early treatment can stop it. Our articles on reversing early decay and on watching a small cavity or filling it go into that decision. The different types of image are explained in our guide to bitewing and OPG X-rays.
Software is being trained on the same problem
A 2026 Malaysian study used 1,410 bitewing X-rays (1,128 for training and 282 for validation) to train an artificial intelligence model to outline decay between the teeth and cervical burnout separately. It performed well on its validation images, but less well on a validation subset that had not been digitally altered (augmented), and the authors called for testing on independent images from broader populations (Ismail and colleagues, 2026). The study did not compare the software with a clinical examination, so software like this may support a dentist's reading but has not been shown to replace the examination.
For dentists: the masking test for a Mach band
When a band of apparent radiolucency sits at the enamel and dentine junction on a bitewing, the 2026 review describes a simple check: visually mask the more radiopaque enamel, for example with an opaque card over the enamel margin. If the band disappears, it is likely to be a Mach band effect; if it persists, it may be a true lesion in the outer dentine. For cervical shadows, note the position relative to the cementoenamel junction and the alveolar crest, and compare with a second projection where one is available, since burnout seen on periapical views of premolars and molars may not appear on a bitewing (Rahmatulla and Wyne, 1995).
The limits: the masking test is described in the review's background and discussion, but none of its five included studies measured how accurate the test is, and we did not find accuracy data for it. The review's evidence was rated low to very low certainty with high heterogeneity, and the burnout distribution data come from a 1995 retrospective study of 25 radiographic series. Treat these as aids to interpretation, not as diagnostic rules, and confirm with clinical findings.
Questions patients ask about X-ray shadows
Why did the shadow not show on my last X-ray?
Optical effects can depend on the angle and type of image. In one study, 70% of the burnout seen on premolars in periapical X-rays was not visible on the bitewing view. A shadow that comes and goes between images is worth discussing with your dentist, who can compare them.
Is a dark area at the neck of the tooth the same as a dark area at the root tip?
No, they are different findings. Our article on a dark area at the root tip covers that finding, and our article on a white patch on an X-ray covers the opposite.
Are children's X-rays harder to read?
For this particular effect, the evidence suggests so, although it is limited. In the studies in the 2026 review, non-decay triangular shadows were more common on baby molars than on permanent molars, and the authors concluded that the risk of a false positive was greater in children.
Sources
- Dioguardi M and colleagues. Radiographic artifacts in the diagnosis of dental caries: systematic review with meta-analysis. Oral Radiology, 2026. PubMed 41331196 (open access)
- Yeung AWK. The diagnostic relevance and interfaces covered by Mach band effect in dentistry: an analysis of the literature. Healthcare (Basel), 2022. PubMed 35455810
- Rahmatulla M, Wyne AH. Classification of cervical burnout and its distribution in the dentition. Indian Journal of Dental Research, 1995. PubMed 9495097
- Schwendicke F, Tzschoppe M, Paris S. Radiographic caries detection: a systematic review and meta-analysis. Journal of Dentistry, 2015. PubMed 25724114
- Ismail MIB and colleagues. YOLOv8m-segmentation for detecting cervical burnout and caries in bitewing radiographs: a deep learning approach. Imaging Science in Dentistry, 2026. PubMed 41928845
- healthdirect. Tooth decay. healthdirect.gov.au
If you have been told about a shadow on an X-ray and would like it looked at, you can book an appointment at our Melrose Park practice or read about general dentistry at Lumi.
This article is general information only and is not a substitute for an in-person assessment by a dental or medical professional.









