You have an MRI booked and the safety form asks about metal in your body. In the studies below, ordinary crowns and fillings were not found to be a safety problem in the scanner. What dental work can change is the picture close to the teeth, and how much it changes depends largely on which material is in your mouth. The one item that needs a specific mention is a denture held in with magnets.
Two separate questions: is it safe, and will the picture be clear
An MRI scanner uses a strong magnetic field and radio waves. A 2018 literature review in Restorative Dentistry and Endodontics (Chockattu and colleagues) described three ways dental materials can interact with it: artefacts that distort the image, heating, and a pull or twist on magnetic metal. Safety is about the last two. Image quality is about the first, and it usually matters most when the area being scanned is the mouth, jaws, sinuses or face.
Most crown metals barely move or warm
A Japanese study in Magnetic Resonance in Medical Sciences (Oriso and colleagues, 2016) tested dental metals in a 7 tesla research scanner, a much stronger field than the 1.5 and 3 tesla scanners used in most of the other studies below. Full metal crowns made from cobalt-chromium and nickel-chromium alloys were deflected by 18 and 13.5 degrees. Crowns made from the other metals tested were not deflected at all. Every crown warmed by less than 1°C in every scanning sequence. The authors concluded that most metallic dental materials showed no apparent attraction or heating at 7 tesla.
All-ceramic crowns were not part of that test. For precious metals such as gold, the studies that looked at them, including a cadaver study of gold-platinum metal-ceramic crowns (Gao, 2022), measured image effects rather than safety. If you have a gold crown or onlay, image quality near the tooth is the question those studies address.
Magnetic denture attachments are the exception
Some dentures clip onto small magnets, with a metal "keeper" fixed to the root or implant underneath. These behave differently. In the same 7 tesla study, the keepers were pulled to 90 degrees or more. A 3 tesla study in Dentomaxillofacial Radiology (Hasegawa and colleagues, 2015) found little heating, but every deflection angle measured exceeded 45 degrees. The authors advised checking magnetic attachments before and after an MRI and removing large prostheses that contain cast magnetic alloy.
If your denture or overdenture uses magnets, say so on the MRI form and bring it up with the radiology staff before the scan. Your dentist can check the keepers afterwards.
Which materials affect the image
Studies have measured artefacts from dental materials in gel models, cadaver heads and patients. The findings line up reasonably well, although a 2022 systematic review in Dentomaxillofacial Radiology (Bohner and colleagues, 19 studies) judged the evidence on crowns, bridges and implants inconclusive, so treat this as a guide rather than a rule.
| Material | What the studies found |
|---|---|
| Zirconia and other all-ceramic crowns | Barely any artefact on a cadaver head at 1.5 T (Gao, 2022). Zirconia was rated fully compatible even for dental MRI (Tymofiyeva, 2013). |
| Porcelain fused to a precious metal (gold-platinum, silver-palladium) | Barely any artefact on a cadaver head (Gao, 2022). |
| Porcelain fused to a non-precious metal (cobalt-chromium, nickel-chromium) | Significant artefacts (Gao, 2022). Cobalt-chromium was classed as not compatible with dental MRI (Tymofiyeva, 2013). |
| Gold alloy, amalgam, titanium | Some distortion: placed in a middle group ("compatible I") in the Tymofiyeva classification, below fully compatible materials such as zirconia. Pure titanium crowns produced significant artefacts on the cadaver head (Gao, 2022). In 94 children aged 3 to 15 who had head MRI scans, amalgam and tooth-coloured fillings were not linked to distortion (Tunlayadechanont, 2024). |
| Stainless steel (preformed metal) crowns on children's baby teeth | The only restoration linked to distortion in the children's study. More crowns meant larger artefacts, but almost all distortion was confined to the mouth and the maxillary sinus and did not affect the brain areas being assessed. |
| Composite fillings | Varies by product. Composites from two makers fell into different groups in the Tymofiyeva classification. In a gel test of 44 dental materials of all kinds (Klinke, 2012), 13 produced MRI artefacts at 1.5 T, and ytterbium trifluoride, found in some composites, was among the causes. |
In these studies, the material a crown is made from mattered more than simply having a crown. The stainless steel crowns used on baby molars are worth listing on a child's MRI form, even though the Thai study found their distortion was almost always confined to the mouth and sinus.

Braces and fixed retainers cause the most trouble
The Bohner review found that orthodontic appliances were the best studied group and that stainless steel brackets were the most likely to cause artefacts. They can affect more than the mouth. A Canadian study at the Hospital for Sick Children in Toronto (Zhylich and colleagues, 2017) placed four kinds of appliance in trays worn by 10 adults during 3 tesla head scans. Steel brackets and molar tubes affected the images the most. Ceramic brackets caused minimal distortion.
A Swedish study (Sonesson and colleagues, 2021) found that conventional steel brackets, nickel-free brackets, a fixed retainer and a palatal expander caused major artefacts in the upper and lower jaw, and steel and nickel-free brackets also caused major artefacts in the eye region at 3 tesla. A scanning sequence designed to reduce metal artefacts improved the images in both scanners.
Bonded retainers vary by wire. In a dry skull study (Shalish and colleagues, 2015), one of two common retainer wires caused no distortion, while the other caused distortion in 46% of tests in areas close to the retainer (tongue and jaws), with more distortion at 3 tesla, with an upper retainer and with retainers on both jaws. The authors concluded the first did not need removing, and that removing the second should be considered when the scan is of areas close to the retainer, particularly at 3 tesla or with retainers on both jaws.
Whether braces or a retainer should come off before a scan is a decision for the doctor who ordered it, the radiologist and your dentist or orthodontist together. It depends on what is being looked for and where.
If you have MRI scans regularly
For someone who has head and neck scans every year or two, the choice of crown material can be part of planning. In the cadaver study, zirconia, ceramic and precious metal-ceramic crowns produced fewer artefacts than non-precious metal-ceramic crowns. Scanner strength and settings also make a difference: the retainer study found more distortion at 3 tesla than at 1.5 tesla, and a 2026 laboratory study of a 0.55 tesla dental MRI system (Sampaio-Oliveira and colleagues) found artefacts from dental materials were mostly confined to the material and its immediate surroundings, with their type and extent depending on the scanning sequence.
Lumi uses e.max (lithium disilicate) and offers zirconia for crowns. If you mention your scanning history at the consultation, it can go into the discussion about ceramic or zirconia crowns alongside strength, appearance and how much tooth is left. Implants raise their own questions, which are covered in our post on dental implants and MRI scans.
What to put on the MRI form
The Better Health Channel's MRI scan page advises telling your doctor about any internal device or implant you have. For your mouth, that means:
- crowns and bridges, and whether they are metal, metal under porcelain, or all-ceramic if you know
- braces, bonded retainers behind the front teeth and palatal expanders
- dentures or overdentures that clip onto magnets
- dental implants
- removable plates, retainers and mouthguards, which you can usually take out
If you are unsure what a crown is made of, your dentist's records or the lab's paperwork may say.
For dentists: writing a pre-MRI note
When a patient asks for a letter before an MRI, the most useful detail is the material or alloy class, which the patient record or lab paperwork may show. Non-precious metal-ceramic crowns produced significant artefacts in a cadaver study (Gao, 2022), cobalt-chromium and stainless steel orthodontic appliances were classed as not compatible with dental MRI (Tymofiyeva, 2013), and cobalt-chromium and nickel-chromium crowns were the only full metal crowns deflected at 7 tesla (Oriso, 2016). Precious metal-ceramic, zirconia and ceramic crowns barely produced artefacts in the cadaver study.
Magnetic attachment keepers are the one item where a safety step is suggested: Hasegawa and colleagues recommended inspecting them before and after the scan. Keep the limits in mind. Most of this evidence comes from phantoms, cadavers and small samples, and the Bohner review called the restorative data inconclusive, so the note should describe what is present rather than promise an artefact-free image.
Questions about dental work and MRI
Do I need my crowns removed before an MRI?
The studies here did not identify ordinary fixed crowns as a safety problem at the field strengths tested. The question is usually image quality, and only when the scan covers the area near the crown. The radiology team can decide whether that matters for your scan.
Will my fillings heat up?
None of the studies here measured heating in fillings on their own. In the 7 tesla study, full metal crowns, which contain more metal than most fillings, warmed by less than 1°C.
Do amalgam fillings show up on a brain scan?
In the study of 94 children, amalgam fillings were not linked to any distortion on head MRI scans.
Talking about crown materials
If you need a crown and have regular scans, or you are not sure what is already in your mouth, a dentist can go through your options with you. Read more about crowns and bridges at Lumi, or book a visit at our Melrose Park practice.
Sources
- Chockattu SJ, Suryakant DB, Thakur S. Unwanted effects due to interactions between dental materials and magnetic resonance imaging: a review of the literature. Restor Dent Endod. 2018. PubMed 30483463
- Oriso K and colleagues. Impact of the static and radiofrequency magnetic fields produced by a 7T MR imager on metallic dental materials. Magn Reson Med Sci. 2016. PubMed 25994037
- Hasegawa M and colleagues. 3-T MRI safety assessments of magnetic dental attachments and castable magnetic alloys. Dentomaxillofac Radiol. 2015. PubMed 25785821
- Bohner L and colleagues. Artifacts in magnetic resonance imaging caused by dental materials: a systematic review. Dentomaxillofac Radiol. 2022. PubMed 35348371
- Gao X, Wan Q, Gao Q. Susceptibility artifacts induced by crowns of different materials with prepared teeth and titanium implants in magnetic resonance imaging. Sci Rep. 2022. PubMed 35013440
- Tymofiyeva O and colleagues. Influence of dental materials on dental MRI. Dentomaxillofac Radiol. 2013. PubMed 23610088
- Tunlayadechanont P and colleagues. Association between dental restorations and artefacts on head magnetic resonance images in paediatric patients. Int J Paediatr Dent. 2024. PubMed 38195814
- Klinke T and colleagues. Artifacts in magnetic resonance imaging and computed tomography caused by dental materials. PLoS One. 2012. PubMed 22384071
- Zhylich D and colleagues. Effects of orthodontic appliances on the diagnostic quality of magnetic resonance images of the head. Am J Orthod Dentofacial Orthop. 2017. PubMed 28257733
- Sonesson M and colleagues. Orthodontic appliances and MR image artefacts: an exploratory in vitro and in vivo study using 1.5-T and 3-T scanners. Imaging Sci Dent. 2021. PubMed 33828963
- Shalish M and colleagues. Influence of common fixed retainers on the diagnostic quality of cranial magnetic resonance images. Am J Orthod Dentofacial Orthop. 2015. PubMed 25919106
- Sampaio-Oliveira M and colleagues. Influence of dental materials on artifacts and image quality in dental-dedicated MRI. Oral Surg Oral Med Oral Pathol Oral Radiol. 2026. PubMed 42288416
- Better Health Channel. MRI scan. betterhealth.vic.gov.au









