MRI And Metal

Many people just learning about MRI safety and hazards ask very similar questions. One of most frequent is, “why do I have to remove all metal before an MRI,” or it’s corollary, “can I get an MRI with some metal on (or in) me?” To answer these questions, let’s start at the very beginning…

As far as the MRI machine is concerned, there are two different types of metal, ferromagnetic and non-ferromagnetic. You may remember back to high school chemistry and the periodic table of elements where many of us learned (and then promptly forgot) that Fe is the symbol for iron.

Fe - Iron from the Periodic Table

Fe - Iron from the Periodic Table

“Fe”, the symbol, is derived from ferrum, the Latin word for iron. Ferromagnetic does not mean that a metal contains iron, but rather that the material has magnetic properties as iron can.

Ferromagnetic metals are iron, cobalt and nickel. These raw ingredients are common in many, many things made from metal, including (likely) the steel grommets in your shoes, to the zipper in your pants, to components in your wristwatch. Another common area to find these metals is in batteries, such as those found in your hearing aid, cell phone and iPod. There are a few non-metal ferromagnetic materials, but these are not very common.

Alright, alright, already… enough chemistry. What does this mean?

When exposed to magnetic fields, ferromagnetic materials become magnets themselves. You can prove this yourself with a fridge-door magnet and a few paper clips. You’ll probably find that paper clips right out of the box aren’t capable of magnetically ‘sticking’ to one another. If you stick one to a chunky fridge-door magnet, however, that paper clip is now magnetized and will likely be able to magnetically ‘stick’ to another paper clip. The length of the magnetic chain of paper clips you can create is a function of how strong the fridge-door magnet is and the magnetic properties of the paper clip steel.

Now, the exact same thing happens with ferromagnetic metals approaching the MRI, but a crucial difference is the distance at which the materials get attracted. With your fridge-door magnet test, the paper clip needs to be touching (or very nearly so) the magnet before the attractive effects are felt. MRI’s, by virtue of the fact that they’re both 1,000’s of time stronger and larger than your fridge-door magnet, can exert profound attractive force at a good distance away from the magnet.

The size and strength of MRI magnets is so great that people have been trapped, injured, and even killed by the force of ferromagnetic objects attracted to the MRI. From concealed roller-skate tennis shoes, to steel-reinforced furniture, to conventional hospital wheelchairs and gurneys, to steel oxygen cylinders, all of these normally harmless (outside the MRI suite) items become life-threatening when subjected to the enormous pull of the MRI’s magnet.

Not all metals are ferromagnetic. In fact, in an MRI suite a concerted effort is usually made to rid the area of ferromagnetic materials and use non-ferromagnetic replacements whenever possible. Non-ferromagnetic metals include aluminum, titanium, brass, copper, and many others. These (and other) non-ferromagnetic metals can present other problems and hazards during MRI imaging, but that’s a topic for another day.

It is almost impossible to determine whether a material is ferromagnetic just by looking at it. In fact, even sometimes when you know what an object is made of, it still isn’t enough to know whether it’s ferromagnetic or not. Stainless steel, is one of these examples.

Stainless steel is not a metal, but rather a family of recipes for metal. Some stainless steel ‘recipes’ (alloys) call for ingredients with ferromagnetic properties. Others which include ferromagnetic ingredients are specially formulated to change the structure of magnetic materials into non-magnetic versions of the material. These special ‘de-magnetized’ stainless steels can become ferromagnetic if the steel is manipulated (shaped, bent, heated, or stressed), so even magnetically ‘safe’ stainless steels can become ‘unsafe’ under certain circumstances (a change that isn’t observable to the eye).

It is remarkably difficult to distinguish magnetically ‘safe’ metals from magnetically ‘unsafe’ metals, either by simply looking at them or, sometimes, even if you know what the metal is. As a result, MRI facilities must assume all metals to be magnetically unsafe unless and until they’ve been verified to be non-magnetic.

So, how do MRI facilities distinguish magnet-unsafe metals? They can use magnets, which shouldn’t be used on patients or sensitive equipment, limiting their applicability. The safer option (and arguably more effective, to boot) is to use a ferromagnetic detector, at least on patients and sensitive equipment.

Ferromagnetic detection instruments, such as the Mednovus products, should be used to help identify magnetically-unsafe materials. This is the standard established by the American College of Radiology, the VA’s MRI Design Guide, and even recommended by the Joint Commission in Sentinel Event Alert #38.

As a patient, it is vital to take seriously the admonitions against wearing or carrying metal into the MRI suite. If you have shrapnel, penetrating metal injuries (particularly in the eye), or any surgeries, implants or prosthetics, it’s critical to have the full information on each to share with your MRI provider. Metal inside the body may not fly across the MRI room like a loose oxygen cylinder (don’t believe what you see on House), but the twisting an pulling that the magnet will exert on an internal ferromagnetic object can be just as dangerous. Active implanted devices, such as pacemakers or nerve stimulators, present particular problems because of both the magnetic attraction and potential interference with the normal function of the device.

Patients should also actively seek out MRI providers that conform with the contemporary safety recommendations, including the use of ferromagnetic detection. You can even contact Mednovus when you want to find providers near you who have this technology available.

Providers of MRI services should make sure that the pre-screening and safety services they provide are in accord with the contemporary best practices, including the use of ferromagnetic detection. With available ferromagnetic detection products equal in cost to only a few hours worth of technical revenue, there’s no financial rationale for not providing this valuable safety benefit to patients and staff. Plus, when weighed against the costs of ferromagnetic object accidents, these instruments of safety are clearly effective risk-management investments.

In all cases, metal brought to the MRI suite (either inside or outside the body of the visitor) should be scrutinized by a trained MRI staff person. This investigation should be aided through the use of ferromagnetic detectors, both to help characterize the hazards of any particular object and to help find ferromagnetic materials that weren’t caught in the prior screening process.

Tobias Gilk, President & MRI Safety Director
Mednovus, Inc.

84 thoughts on “MRI And Metal

  1. Tobias Gilk Post author


    I would think that pretty much any plastic would be fine.

    I wish you well.


  2. Mary

    Is there a specialized MRI that can detect and/or minimize artifact from stainless steel? I have an ankle injury which had a stainless steel plate and screws but need an MRI for diagnosis of tendon damage but was told that no MRI can be done because of the metal artifact.

  3. Tobias Gilk Post author

    Mary, thank you for your question. Some MRI scanners do have software that allows for special “metal suppression” imaging sequences. These can improve image quality (reducing artifact / distortion from metal materials), but they can’t totally eliminate the negative effects of metal. The specific type and grade of the stainless steel implant will also be a determining factor in how much the metal suppression techniques are effective. Irrespective of the grade of stainless steel, if the implant is very close to the tendon that your physician wants the image of it may not be feasible.

    To find out if it may be possible, I would suggest contacting the radiology department of the largest hospital around, or a hospital associated with a medical school. Ask to speak with the most senior MRI technologist, and explain to him / her your situation. Ask about their capability to perform metal suppression sequences. If they don’t have the right MRI scanner (or software options for the MRI scanner), ask them who they think might, then call them.

    The technologist may tell you that the plate and screws are just too close to the tendon that the doctor wants to see, and that no contemporary MRI will be able to get useful images that close. I very much hope that this isn’t the case for you.

    Best of luck to you!

  4. Mary

    Is there a specialized MRI that can minimize the artifact effect from stainless steel? I have 6 screws and a plate in my ankle but need an MRI to diagnose tendon and ligament problems, the hardware cannot be removed for another 7 months.

  5. Tobias Gilk Post author

    In general, you want to find a lower-field-strength MRI system (the less the magnetic field, the less artifact). And like your personal computer, it’s the software that makes the magic happen. The software version and options that the provider has will make a significant difference. There are generic settings that any MRI operator can use (see, or, and there are more effective proprietary software systems, with acronym-rich names that vary depending on the system manufacturer (MARS, MAVRIC, SEMAC).

    The only way to sort through these variables will be to talk with the MRI provider, and let them evaluate the options based on your implant, the anatomy that needs imaging, and the capabilities of the hardware and the software of the MRI center.

    I hope this helps.

  6. Natalie

    I recently had an MRI and part of the images could not be seen, due to a “metal object.” Trouble is…I have NO metal at all in my body. Have you ever come across such an instance? I’m bewildered and appreciate your input. Thanks!

  7. Tobias Gilk Post author

    Even if you have no metal in your body, that doesn’t necessarily rule out the possibility of there having been metal in the MRI scanner with you. While they’re not very common, there are personal care products (lotions, deodorants, haircare products) that can have metals in them. Also there are increasingly clothing items that are ‘anti-microbial’ and are impregnated with metallic nano-particles. Perhaps one (or more) of these items contributed to the effect you experienced.

    I hope this is helpful.


  8. Nicholas

    However with the new, more powerful MRIs coming out is it possible that even “non-magnetic” metals like copper and aluminum could be a safety hazard? For example, I see that there is now an 11.75 tesla MRI now and I know copper IUDs are only considered safe for up to 3 tesla. The strongest magnet they have, which has not been used in a MRI, is 70 telsa. I am sure that would have some nasty effects on copper.

  9. Tobias Gilk Post author

    Nicholas, pretty much everything ‘can’ become magnetic, if you apply enough external magnetism to it… but even at 11.75 tesla, copper is going to behave pretty much like we expect it to outside of that magnetic field.
    Now, all metals can be subject to heating effects, and the risk of heating is tied to the size of the object and the field strength (really, the imaging frequency, which varies with field strength). At higher and higher field strengths, the size of an object with the greatest heating risk gets smaller and smaller. For example, at 3T, a straight electrical conductor inside a person’s body is at the greatest risk of heating if it’s about 15 cm in length. At 11.75T, the greatest risk would be a bit shorter than 4 cm. So objects that might be safe from significant heating at 3T, might be at great risk of heating at 11.75T. And here’s the interesting thing… the same is true in reverse… objects that might be at great risk of heating at 3T, might be safe from thermal effects at 11.75.
    While you can apply backwards-compatibility for magnetic attraction and force (if it’s safe for attraction effects at 3T, it is necessarily safe at lower field strengths), the same can not be said for RF thermal risks.
    I hope this helps.

  10. Martin

    Hello ,I have been told I need shoulder replacement ,I live with a lot of pain. It’s getting worst not better,I recently had an ( MRI ) and they discovered ferragmentic material in shoulder joint.what does this mean ? It looks like a be be stuck into the joint or a pellet. Can anyone anwser my question ? Thanks .
    Sore shoulder !

  11. Terri

    Nickle dental instruments broke off in my mouth during a root canal. Would that be a problem for an MRI of head and neck?

    My sister needs an MRI of her brain, but is considering a procedure to add gold to her eyelid to weight it down and help it to close at night (effect of a cranial nerve schwannoma). Would this procedure affect her MRI results or her safety during the MRI?

    Many thanks,

  12. Peggy

    Hi Tobias,
    I had an MRI today. I have 18 facial piercings. Before the test I made sure I had no metal in/on any of my clothes and the I proceeded to take out my 14 earrings, 1nose ring, and 2 lip rings. In my haste and nervousness (they are testing me to see if I have MS at age 25) I forgot about my tongue ring. It is stainless steel. And nothing happened during the test. I didn’t even realize I had forgotten it until I started putting the rest of my rings back in place. Now I’m worried that it may have messed up the images.
    Thank you for your time,

  13. Leonora

    In 1986 a TA55 stapler was used on my colon. Can I have an MRI? In this case on my Rotator cuff and/or my knee?

  14. Tobias Gilk Post author

    It’s not uncommon for adults to find that there’s a foreign body, sometimes metallic, inside their bodies. The significance of this (from an MRI safety point of view) depends on whether the material is magnetic (which it sounds as if someone has already determined it is), the object’s mass and shape (bigger, or longer, or sharper are worse), and what structures the object is near.

    An MRI safety-trained radiologist should be able to help you answer each of these questions, perhaps with the benefit of a CT exam to more accurately characterize the foreign body in your shoulder.

    I wish you well…

  15. Tobias Gilk Post author

    Metal inside the body will disrupt an MRI image of the anatomy immediately adjacent to the metal. That disruptive ‘artifact’ varies in reach based on the size and magnetic properties of the metal involved.

    Nickel is a strongly ferromagnetic metal, by itself, though it can be formulated into alloys that have virtually no magnetic action. If there’s a fragment of a dental instrument in your mouth, I would operate from the assumptions that (a) it is ferromagnetic, and (b) it’s probably fairly small. For many MRI imaging techniques, it’s likely that the artifact from this sort of metallic foreign body would be measured in millimeters. There are some types of MRI imaging that can be extraordinarily sensitive to metallic foreign bodies, where the artifact could be measured in centimeters.

    For your sister, the consideration would be similar for the eyelid weight, except that gold has far less magnetic ‘action’ to it than does nickel. This would mean that the artifact associated with a gold foreign body should be significantly less than the artifact associated with a nickel foreign body (all other things being equal).

    I hope this helps.

  16. Tobias Gilk Post author


    The good news is that technologist / radiographer who was running the MRI scanner is reviewing the images as they’re being generated. Yes, a stainless steel tongue post likely did disrupt the image, but that disruption or ‘artifact’ may well have been limited to the area of your mouth, and not negatively impacted the imaging of your brain. If the technologist / radiographer had seen otherwise, I suspect that they would have asked you if there was anything in your mouth before completing the exam.

    I hope this helps.

  17. Tobias Gilk Post author

    The answer to this may depend on the specific surgical staples that were used. If possible, get a copy of the medical records from that procedure to identify the specific staples (type, material, quantity) and provide them to whoever would perform your MRI so that they can evaluate the MRI safety of your implants.

    I hope this helps.

  18. Rick Kennedy

    Hi Tobias,

    I’m working with our MRI department to streamline some of their processes. We spend a lot of time verifying implant safety. Many times, patients don’t know their implant details, don’t carry their card, and/or are incapacitated. Families often don’t know, either. This leads to a lot of delays and investigative work. Is there technology that can detect MRI-unsafe implants?



  19. Ed

    I have a nerve stimulator implanted in my back. The problem is I need to have an MRI, but because of the stimulator I cannot. What else is can be done with close to the same results as an MRI? I have already had a CT scan. I was told that I need an MRI for better results than the CT scan. Any help is greatly appreciated.

  20. Kristie

    Hi Tobias,
    I was wondering if you could give me your opinion. When I was a toddler I had open heart surgery in 1987. I had an atrial septic defect that was closed successfully. A sternotomy was performed. Reading my records, it says they placed pacing wires and no. Interrupted stainless steel wires. My neurologist would like me to get an MRI of my brain. My mother was told at the time of my surgery that I could never have an MRI done. A radiologist looked at one of my past chest x-Rays and said that was enough to determine that it’s safe for me to have an MRI. I have since gotten the opinion of two other radiologists that have all said stainless steel is not magnetic and that it has been there for so long that it would not move. I’m still really nervous and wonder how to proceed. They said once I go in there is really no further screening to determine if the metal is ferromagnetic. Could this be unsafe? What else can I do to determine that having this done won’t heat the metal in my chest or move it? Would a ferromagnetic detection scanner prove that this is safe? Any thoughts on this would be very appreciated! Thanks!

  21. Tobias Gilk Post author


    You have hit upon one of the most time-consuming parts of screening patients. In a highly unscientific poll, I found that an average amount of time spent researching implants / devices is about 5 hours per week per MRI. So if your site has three MRI scanners, on average you’d be spending about 15 hours per week chasing down implant data. If you’re in an academic medical center (or other site that does other urgent or high-acuity patients), that number may be closer to 8 hours per MRI.

    There is not any patient-screening technology (of which I’m aware) that will speed this process. There is, however, a process that can streamline the risk-assessment portion of this, and can make a significant dent in the amount of time spent.

    Dr. Emanuel Kanal teaches an MRI safety course ( in which he details a review process that can streamline the process. At one site that I know of that was around 8 hours per magnet before this training, they cut the time spent doing patient screenings to nearly 1/4 of what it had been (down to about 2 hours per magnet). This process does require some training of techs and radiologists, and it does require that department management be ‘on board’.

    I suspect that wasn’t the answer that you were looking for, but it is the closest thing to a ‘screening silver bullet’ that I’ve ever come across.

    I hope this helps.


  22. Tobias Gilk Post author


    As your doctors may well have told you, there are qualitative differences between what a CT can show, and what an MRI can show. Depending on the make and model of your nerve stimulator, it may be possible to have an MRI. I would recommend that you get as much detail about the stimulator as you can (manufacturer, model, serial number, date of implantation), and get it to the radiologist at the location where you would have this MRI done (for this, it might be best to have it done at a hospital, as opposed to an outpatient imaging center). They may be able to figure out a way to safely provide you with an MRI exam, but it will be contingent upon what the stimulator is, and where it is positioned relative to where they want to do the MRI exam.

    I hope this helps.

    Tobias Gilk

  23. Tobias Gilk Post author


    That the wire is identified as stainless steel does not necessarily mean that it is non-magnetic, but think about it this way…

    If I had an extremely large, extremely powerful magnet, and put a paperclip on it, you’d still be able to pluck the paperclip off of the magnet. This is because the attractive force of a magnetically drawn paperclip is not only a function of the strength of the magnet, but also of the mass of the paperclip. You know, intuitively, that the paperclip would be easier to remove than, say a 5 lb iron dumbbell.

    It would be my presumption that the wire used to tie the sternum of a toddler together would be fairly fine, and that fine wire (aka, relatively low mass for unit length) is now woven into the bone and cartilage of an adult. If it pulls, it would be pretty well anchored. I, personally, would not be concerned about significant risks of attractive force acting on the wire if I were in your shoes.

    The other issue that you raised is that of heating. MRIs use radio frequency energy as an element to create the images. That RF energy naturally gets converted into heat inside the MRI patient’s body (a low-level gentle warming). Metal objects, however, if they are bathed in RF energy have the potential to act as antennae and ‘suck up’ the RF energy and concentrate it. Now, the ability of a metal object to act as an antennae is directly related to it’s size, shape, and orientation. There are two ways that this could be addressed for the MRI of the brain, such as you described.

    Each MRI field strength (1 tesla, 1.5 tesla, 3.0 tesla) uses a different RF frequency, and each frequency has a different ‘antennae length’ that is worst-case for heating. 3.0 tesla MRI scanners have a ‘worst-case’ length of about 15 cm. 1.5 tesla MRI scanners have a ‘worst-case’ length of about 30 cm. And as field strengths get lower, the ‘worst-case’ length gets longer and longer. As long as the field strength (frequency) being used for your MRI isn’t close to the effective length of the wire (and the radiologist may want to look at an x-ray of your chest and figure out what the effective length is), then, in my opinion, the risk of heating of the wire is quite low. So it is conceivable that the length of the wire is such that your radiologist believes that it wouldn’t be of any significant concern.

    The other way to address risks of heating for your brain MRI would simply be to not put any RF into your chest. MRI’s usually transmit the RF from the “walls of the tube”. This means that a large part of your body might get the RF energies. If you were having your brain imaged – depending on the MRI scanner being used – you might also get a similar level of RF energy deposited in your chest area. If, however, your MRI provider has what is called a transmit-receive head coil, which enables the RF transmit to come from the ‘cage-looking-thing’ that they put around your head, instead of from the walls of the tube, then nearly all the RF energy for the study goes to what they’re taking the pictures of (your head), and almost none of it goes anywhere else.

    I hope this helps, and that you and your doctors are able to find a safe and effective way for you to get the MRI scan that they think would be helpful for you.

    Tobias Gilk

  24. Kira

    Wow! Great advice! I’m due to have a MRI scan in a few weeks, but just had hair extensions done. They are put on with kind of metal rings,and can’t be removed temporarily. If I take them off, I would have lost so much money. They will last for 4 months.
    The MRI scan would be of my neck. Would my hair extensions be too close to interfere, or could it cause me some damage?

    Many thanks in advance. Have a good day!

  25. Tobias Gilk Post author


    Generally speaking, extensions with metal clips, rings, or wires will create artifacts in the general vicinity of the metal… generally within a centimeter or two. For some, very sensitive types of scans, that distance can extend. I suspect it won’t be a concern if you push all of your hair above the top of your head when you’re getting your neck imaged, but I would strongly recommend that you contact the MRI center where you’re having the scan to verify with them that it will be OK.

    I hope this helps.


  26. Crystal

    When working with metals such as metal fabrications what affect can it have in the body after an MRI

    My husband has worked many years with metals, welding and grinding etc. Since his recent MRI he is experiencing severe joint pain, blurred vision, chilling then hot, a hit spot in his head.
    This is becoming seriously uncomfortable for him
    What can be done to ease his pain etc

    Thank you in advance

  27. Tobias Gilk Post author


    I am not personally aware of any correlation between a history of metal work and any MRI risks, other than the immediate ones (for example, retained metal within the body being pulled / twisted by magnetic force). I’m not aware of any mechanism that would link the longer-term symptoms you have reported with just a non-contrast MRI exam. I would strongly recommend that he see a doctor to try and identify what the cause(s) of these symptoms are.

    I hope this helps.

  28. DeLos Allen Bates Jr

    I have a stainless steel rod in both lower legs thanks to a drunk Korean woman running me down while I was riding my bike on the sidewalk in 1993. I also have 3 screws in each rod. In 2009 I had an MRI done in El Paso, Tx and was given the little 2-page Questionnaire to answer. When they did the MRI they placed lead aprons across my lower legs as a satefy measure. Last year, December 2014, they did another MRI on my lower Lumbar, but this time I wasn’t handed a Questionnaire or even asked any of the questions here at Denver Health Medical Center in Denver, Co., as a matter of fact, they filled out both pages themselves and answered “No” on every question. They didn’t even place the lead aprons across my legs. What I want to know is do I have a Lawsuit against them for failing to follow Standard Operating Procedures?

  29. Tobias Gilk Post author

    You’ve raised a few different points here.

    First, MRI does not make use of any ionizing radiation (which is what lead shields protect from), so if the El Paso site actually put a lead apron over your legs, *that* would suggest to me that the people running that site did not know what they were doing. Lead would offer you zero safety benefit and – as an electrical conductor (albeit a poor one) – would have actually introduced the potential for MR-related burns.

    If the Denver site completed the questionnaire by asking you the questions (and they just filled it out), that’s fine. The critical part is that the information on the form is accurate. But if they just completed the form, just to have a completed form, and they didn’t actually complete it with information from your medical history and have you verify it, then that’s a pretty egregious violation of best practices. The entire point of having a screening form is to collect and document the MRI patient’s specific history and potential contraindications, not to simply have a piece of paper with which to stuff a file.

    With regard to a lawsuit, generally one needs to demonstrate damages (injury, incapacitation, loss of money / earnings / or other value) to have a claim, unless the suit is seeking relief in the form of actions that the healthcare provider is compelled to do (or stop doing).

    It is important to note that I am neither a physician, nor an attorney, so my opinions should not be misinterpreted as either medical or legal advice. For those, you should absolutely consult with your physician and attorney (respectively).

    I hope this helps.

  30. Kristie

    Hi Tobias,
    Thank you so much for the information you provided me with. I finally went through with the test. The technologist was great and used the cage-looking thing without me even saying anything. I greatly appreciate you putting me at ease to go forward with the MRI!!

  31. Tobias Gilk Post author

    I’m very pleased that your MRI went off smoothly, and that you felt comfortable during it. I hope that the exam also brought you good news. Thank you for your follow up, Kristie.

Leave a Reply

Your email address will not be published. Required fields are marked *

Change the CAPTCHA codeSpeak the CAPTCHA code