About MRI Safety

MRI is one of the safest medical imaging procedures available.

Unlike X-ray imaging or CT-scanners (which use X-rays), MRI’s do not use ionizing radiation, a type of radiation which causes genetic damage and leads to increased risks of cancer.

MRI imaging, by contrast, uses magnetism and radio frequency (RF) energy, plus some very clever advanced physics, to generate images. To date there have been no quality studies that demonstrate that either magnetic fields or properly administered RF energy have any dangerous long term effects on our biological bodies. This is not to say that MRI imaging is perfectly safe and ‘foolproof.’

The magnets needed for MRI imaging are typically tens of thousands of times more powerful than the Earth’s own magnetic field. These uber-powerful magnetic fields can act on magnetic materials nearby (as in things you carry with you, like a nail clipper or tool) and turn those objects into magnet-homing missiles. The same magnetic forces also penetrate your body and can act equally powerfully on magnetic objects within you (such as aneurysm clips, or pacemakers, or orthopedic implants).

For these reasons it is critically important that every person and object entering the MRI area be comprehensively screened for magnetic / metal materials before they’re allowed to approach the MRI’s giant magnet.

There are other safety concerns attributable to the administration of the MRI exam, itself… things such as the proper settings for the scan, positioning the patient correctly so that they don’t rub up against the scanner, and making sure that the patient has appropriate hearing protection (these scans can be extraordinarily loud).

42 thoughts on “About MRI Safety

  1. Joe

    What are the safety laws, for being called out for an emergancy in- patient at an off site facility for MRI? You are the only employee with an emt, that has no training in MRI. In West Virginia.

  2. Tobias Gilk Post author

    If the patient is currently admitted (sometimes they discharge to send off-site, then re-admit to work around this), then the facility must conform to hospital regulations. If it is JCAHO accredited, then it must conform to those accreditation standards. If the patient is a Medicare / Medicaid patient, then the site must also conform to CMS’s ‘conditions of participation’ requirements, including a requirement to have a licensed physician on site if contrast is being administered. I can’t quote you from these different standards. Generally, they’re all quite weak on MRI safety specifics, but they might have something on emergency preparedness for code situations. Hope that’s at least something…

  3. joon kim

    planning to put an MRI in medical office and need to know the minimal space requirements. We have designated about 600sq feet of space for control room and MRI machine. Is this enough space?

  4. Tobias Gilk Post author

    Good question! The answer, however, depends on a few factors…

    What kind of MRI is it? How big is the machine itself? How far does the fringing field reach? What are the adjacent functions and equipment (including above and below)? Will you support any special clinical functions (e.g. sedation / G.A., image-guided biopsy…)? Will you support any special patient populations (e.g. bariatric, geriatric, pediatric, prisoners…)?

    All of these elements have direct bearing on the size of the room you need.

    Generally speaking, for contemporary, actively-shielded bore-format MR systems, we design the magnet room 18 to 20 feet wide (wider if the room is to support G.A. cases or biopsy), and 25 to 30 feet long (depending on how much of the fringe field we want to capture). This means that the magnet room, alone, will be a minimum of 500 square feet. Computer equipment rooms usually need to be somewhere between 150 and 200 square feet. Control rooms generally get very cramped if they’re less than 150 square feet. That gives you a minimum of 800 square feet, and that is only the interior area of the rooms (add area for wall thicknesses, as well as circulation among the functional areas).

    I hope this helps.


  5. Tobias Gilk Post author

    There are several temporary effects on the body, from the elemental physics changes that enable the imaging to incidental effects such as ‘vertigo’ that may happen. RF energies used during MR imaging may also warm the body of the patient.

    There are not – to date – reproduced studies that show long-term effects from the MR exam, itself, on people who receive them. There is a small (but growing) body of research that indicates that there may be long-term effects of the use of some MRI contrast agents other than Nephrogenic Systemic Fibrosis (NSF).


  6. Alex McCord

    Mr. Gilk,
    I have read the Guidance Document (white paper, 2007 Guidance Document and 2013 Guidance Document). I am an architect working for the Mississippi State Department of Health, and I am reviewing plans for a new hospital. They are employing the zones, but I am working to strengthen their Zone II. I am understanding the intent of Zone II to be a staffed location that prevents access to Zone III prior to screening, etc. The descriptions of the zones don’t use the word “sequential” nor does it necessarily say that Zone II is a staffed location although procedures there would be performed by staff. But here is the nugget of my query. As one enters Zone II, they enter into a corridor. Screening is to one side of the corridor, and a sub-waiting room with dressing rooms is on the other side. The rub I have is that the corridor continues to the entrance of Zone III located further down, so I see a potential breach. Could you give me any further guidance into the intended nature of Zone II and its relationship with Zone III. My email is above there, and I can provide more detail into the particular scenario. Many thanks to you. -Alex

  7. Tobias Gilk Post author

    Mr. McCord,

    You might find an appendix to the ACR Guidance Document (in the 2013 edition, it is appendix 3, “MR Facility Safety Design Guidelines”) helpful to you, as it describes the sequential nature of the Zones.

    In general, Zone II is an area with no immediate or potential MR-related risks. It is the ‘transition zone’ from the wide world (Zone I), to areas where there are either direct or proximal opportunities for MR-related harm (Zone III). Because there are no MR-related risks in Zone II, there is no MR-related reason for continuous observation of persons in this zone. Staffing decisions for Zone II are then relegated to other concerns or considerations.

    Now, once you pass from Zone II to Zone III, now the person is in an area of direct or proximal risk of MR-related harm. For this reason, Zones III and IV require staffing by Level 2 trained personnel to manage the safety of any untrained individuals who are within the controlled access areas of the MRI suite.

    Please let me know if this doesn’t fully answer your questions.

  8. Anna

    I am an ICU nurse and I take my patients to MRI regularly. While I am not in the room while my patient is being scanned, I do enter the MRI suite to get my patient settled on the table.

    I am having a consultation to get hair extensions. The extensions are connected to my real hair with a small metal bead that is made of aluminum/copper composite material. I need to know if this is MRI safe. They are not readily removable (as a bobby pin or a watch would be) such that I could take them out before I enter the room. So I either have to be all in or all out. Once they are in they are in. If a radiologist can let me know the answer I would greatly appreciate it!

  9. Tobias Gilk Post author


    I’m not a physician, but I hope this helps you get to an answer…

    If you’re not going to be receiving an MRI exam, or remaining in the room during an MRI exam, then the only MRI concern / consideration for going into the room is the static magnetic field. If the beads / wire used for the extensions is not magnetic (you could bring a permanent magnet to verify this, for certain, before they get put in), then there are no MRI safety concerns with simply being in the room. If they (beads or wire) were magnetic, you might feel some pulling on your hair. The size / mass of these would still be quite small, and would not likely exert much force. In my opinion, it is very unlikely that anything like the beads / wires getting pulled out of your hair would ever happen.

    I hope this helps.

  10. Jill Schedel

    I’m an MRI tech on a 1.5T magnet … I’m post unsuccessful L Spine surgery and was wondering if I could still work while wearing a morphine pump and/or a neurostimulator. I realize the obvious answer that may enter your mind but I believe their are new and upcoming devices that are conditional. Also any suggestions if I am updated to a 3T … Thanks so much!

  11. Tobias Gilk Post author

    There may be neurostimulators that are MR Conditional. And as a tech and not a patient, you won’t be inside the bore (even if you remained inside the magnet room during scanning), which means that the only MR conditions that would be applicable in that scenario (worker, not patient) would be the static magnetic field conditions (field strength and magnetic spatial gradient limitations).

    The same would be true of a pump, too.

    Without knowing what your specific devices are, there’s no way to offer an opinion on the MR safety your specific devices.

    I hope this helps.

  12. Tobias Gilk Post author

    To answer your question I would need to have some understanding of what these products contain. The website that I found didn’t provide any real information.

  13. Ramin Homanfar D.D.S.

    Hi Tobias
    We are currently developing a technology that will make accessing Central Venous Ports highly precise, reproducible, and easy to perform. Part of the Armamentarium will be the use of Copper wiring. As I understand, Copper is a Non-ferrous material; Nevertheless, you mentioned in your article there may be other concerns regarding the use of some of the materials. I was hoping for an elaboration on the subject.

  14. Tobias Gilk Post author

    Any electrically conductive material has a potential to heat from exposure to / interaction with RF signal that is an integral part of MRI imaging. Copper is an excellent electrical conductor, and at particular lengths and orientations, could be a very significant risk for heating in MRI. I invite you to contact me (tgilk @ MRIpatientsafety . com) if I can help you further.

  15. Luigi S

    I need to demonstrate that some material are “mr safe” following the ASTM;
    – UHMWPE
    – Al2O3
    – PMMA
    I need to demonstrate that this material are:
    – electrically nonconductive
    – nonmetallic
    – nonmagnetic

    Any suggest
    Eng Luigi Schavello

  16. Tobias Gilk Post author


    It sounds as though you need implant design consultation for MR safety. This draws in a number of standards, regulatory bodies, testing methodologies, and more. It’s not an easily answered question.

    I would encourage you to take a look at my consulting services website for information on what can be done on a consulting basis.

    I hope this helps you.

  17. Luca G.

    Someone above asked about hair fibers by Toppik, and I have the same question. The ingredients are: Keratin, Ammonium Chloride, Silica, Water (Aqua, EAU), DMDM Hydantoin, [+/-: CI 15985, CI 19140, CI 45380, CI 17200, CI 42090, CI 61570, CI 42053, CI 61565, CI 15510]

    Thank you.

  18. Theresa

    I resently had a CT scan. Thought I didn’t have any thing on me.
    Late that night after scan I realized I had two safety pins on me.
    Will that hurt me now?

  19. Fred


    I understand that all portable/mobile equipment in an MRI environment needs to be non-ferrous, but was wondering what the situation is regarding fixed equipment – e.g something that is permanently bolted to the wall of the room? Are there restrictions on this also?

    Many thanks

  20. Tobias Gilk Post author


    The overt chemical names don’t appear to me to present risks of significant artifact, but I don’t know what the 9 different compound codes represent, so I can’t give a definitive answer one way or another.


  21. Tobias Gilk Post author


    My area of expertise is really in MRI and not CT, but from my comparatively limited knowledge, I’m not aware of any potential risk / hazard from having had a safety pin (or two) with you during a CT exam.

    I hope this helps.


  22. Tobias Gilk Post author


    It is my personal recommendation to minimize the amount of ferrous material within the MRI scanner room (including fixed), and to choose products / assemblies specifically to reduce the number of tools / devices needed to repair or service. For example, if you had a hand sink in the MRI scanner room, I would recommend using plastic pipe, where practical, and hand-fittings (that avoided the necessity of wrenches) when possible… even if that meant using a metal part (which presumably will have a lower mass than a pipe wrench).

    When looking at bolting *anything* inside the MRI scanner room, be very aware of the construction of the RF shield (Faraday cage). A pinhole in the shield material could negatively affect the image quality of a million-dollar (or more) MRI scanner. If you need to bolt something, make certain that the plan to affix it to a wall / floor takes into account the RF shield integrity.

    I hope this helps.

    Tobias Gilk

  23. Laura L

    Hello, I just had an breast MRI today. I realized after that the pants they gave me to wear had a safety pin on it! The tech was shocked and said it shouldn’t be there. I asked it it would interfere with the images and she hesitantly said It shouldn’t. Could the safety pin have interfered with the ability to get good images and/or was it dangerous?

    Thank you!

  24. Becky

    I had a chemical peel over a year ago. Two weeks ago I had an MRI on my neck, without contrast. That night I had a small sore in corner of right eye. It looked like the beginning of a stye. Area around eye became red and swollen. Redness continued down cheek then it all started on right side of face. I looked like I had a chemical peel again. So many people who saw me thought I had a peel again. I cannot find anything that shows these two have any relation together. Thank you.

  25. Tobias Gilk Post author


    If the safety pin didn’t move, then it’s mere presence was not a safety risk to you. Yes, metal will distort the MRI image of the area right around it, but we’re typically talking about millimeters or centimeters. If the safety pin was on the pants, in my opinion it would not produce a significant distortion to images from a breast MRI.

    I hope this helps.

  26. Tobias Gilk Post author


    I’m not aware of anything that would provide a direct connection between a non-contrast MRI and skin irritation. If you have ongoing health concerns, I would advise bringing them to the attention of your regular physician.

    I hope this helps.

  27. Tara White

    hello i have a maryland bridge on my front teeth and was wondering if the metal used in the bridge will impact my pelvic MRI ?

  28. TJ

    A mobile MRI trailer just moved into my neighborhood within the last week. It is parked at the hospital across the street from where we live in a busy residential neighborhood. I am not imagining it that I can feel when the MRI unit is on, inside my house. I can feel the vibrations of the unit inside my ears, and a cycle of humming comes from the unit all day and all night long. I don’t know what the low humming noise I hear is, or why I can feel the sounds of the unit inside my ears. The feeling is making me nauseas. I haven’t been able to find any information on the internet about living near a mobile MRI unit. Can you please tell me where I can look for information? I am very concerned about the health effects of living near such a unit.

  29. Rick Carpenter

    Hi Tobias,
    Thanks for all the information you’re providing here. I broke my back (L4/L5) in 1972 and had an 8-piece stainless steel cage constructed around my spine that’s still there today. I’ve had recent pains down my legs, and the hospital radiologist (having seen only an X-ray of my back) says I’m safe to have an MRI of my back. Do you have any thoughts on this ?


  30. Tobias Gilk Post author


    In the USA, most (but certainly not all) metals used for orthodontic work are not significantly ferromagnetic. Being fairly superficial, it may be the case that your bridge could be tested for ferromagnetism with either a permanent magnet or a ferromagnetic detector.

    Even if the bridge is significantly ferromagnetic, as long as it is firmly anchored, in my opinion I would not anticipate any problems, except possibly for some mild discomfort. Any distortion of the MRI images would likely be limited to within a few centimeters of the orthodontic appliance, and likely won’t have any impact on the diagnostic quality of an abdominal image.

    I hope this helps.

  31. Tobias Gilk Post author


    For most contemporary MRIs, the magnetic field is on 24/7. If you’re feeling something only during operational hours, it’s not the magnetic field. And if it’s across the street, the amount of magnetism that you’d be exposed to is likely less than what you are exposed to from the Earth’s own magnetic field.

    MRI’s are noisy, however, when they’re in operation. You may be perceiving low frequency sound, though at distance I would expect these sounds to be negligible.

    The MRI may have a generator, and likely have an air conditioning unit that runs all the time. You may be picking up on sounds and / or vibrations from either or both of these.

    Were I in your position, I wouldn’t have concerns about my health from this.

    I hope this helps.

  32. John Bowler

    I am posting this to let everyone know what these machines are capable of. I sustained an RF burn injury during a routine MRI procedure last November. The incident occurred on a Siemens Vero 3Tesla MRI machine during a CSpine/Thoracic scan. Here is a brief account of what happened : When I arrived I noticed the room and MRI bed were warmer than usual because they were very busy and moving people in and out. I was positioned lying on my back and entered head first. There was no padding used during the study but no part of my body touched the walls of the machine. I held the ‘panic’ button in my right hand and my hands came together just below my sternum in a typical ‘praying’ position so my arms rested on my upper abdomen as opposed to down by my sides. The procedure was started immediately. About 10 minutes into the 2nd procedure (thoracic) I started to feel a sort of prickly vibrating heat hitting my skin. It felt like a kind of rippling wave and I started to feel quite warm. It even seemed to vibrate my T-shirt and penetrate through. It was accompanied by a loud hum which I could hear over and above the typical clicking sound. I had never felt or heard anything like this on previous MRI procedures. The sensation wasn’t excruciatingly painful like you would expect from a contact burn, so I decided I would just see it through. I believe the combined CSpine/Thoracic procedure took approximately 30-40 minutes and I felt this sensation for about the last 10-15.

    When I got off the table I felt hot and mentioned it to the technician. She said it was normal to feel a bit warm when the machine had been running for a while. As the day wore on my skin got redder. I felt and looked like I had a bad sunburn. I started applying Aloe Vera gels and lotions. By the next morning I was very red and sore. I was pretty concerned and worried there may be deeper tissue damage. The next day I went to see my doctor. She said she was unfamiliar with MRI related burn injuries but after looking at me felt it had been caused by some type of thermal exposure and advised use of cooling gels and lotions. She consulted with an MRI physicist about the problem and was told : “Regarding the burn: I’d definitely recommend the pt to be checked out by a Dermatologist. MR burns typically starts at subcutaneous fat (no pain receptors) and moves up to epidermis. If the pt was not sedated during the scan and ended up with a burn, there is a clear chance that the patient has damage under the skin that is not visible. I reached out to a Professor of Radiology at USC who looked at my story and pictures and told me this was an RF radiation burn and that I needed to be checked out by a dermatologist ASAP. This was the beginning of a nightmare that has now lasted over 6 months. I have consulted with numerous dermatologists, general practitioners and MRI specialists over this time as well as conducting my own research and have learned a lot about RF frequency injuries and SAR. One of the other MRI physicists I contacted recommended I look at the SAR readings for my tests. He said these are used as an indication of over-exposure and are usually less than 1.

    I was able to find these readings in the DICOM data on the disk (shown below). The overheating feeling I experienced started about 5-10 minutes into the Thoracic procedure which coincides exactly with the 4th thoracic sequence. I was between 2.5 and 2.72 SAR for about 7 minutes and this is when the burning occurred. These SAR reading are of course the machine estimates and not the actual SAR on my body. My belief is that sustained exposure to SAR above 2.5 was too much for my body to dissipate the heat and I burned. Having had many MRI’s in the past I wondered why I had never had any problems before, but when I looked back at the SAR readings for numerous previous exams I noticed it was never more than 1.5 and whenever a sequence showed a higher SAR it was followed by one with lower SAR etc.

    CSpine Sequence start 14:05 End 14:19

    Start DICOM SAR
    1 14:05:17 (0018,1316) DS [0.08906601] # 10, 1 SAR
    2 14:07:43 (0018,1316) DS [0.9164475] # 10, 1 SAR
    3 14:09:52 (0018,1316) DS [1.062324] # 8, 1 SAR
    4 14:12:48 (0018,1316) DS [1.062324] # 8, 1 SAR
    5 14:15:28 (0018,1316) DS [0.07535206] # 10, 1 SAR
    6 14:18:50 (0018,1316) DS [0.7580933] # 10, 1 SAR
    => Thoracic Spine Sequence Start 14:19 End 14:37

    Start DICOM SAR
    1 14:19:29 (0018,1316) DS [1.7210823] # 10, 1 SAR
    2 14:21:36 (0018,1316) DS [0.07548546] # 10, 1 SAR
    3 14:22:24 (0018,1316) DS [0.15613435] # 10, 1 SAR
    4 14:24:37 (0018,1316) DS [2.560578] # 8, 1 SAR
    5 14:27:15 (0018,1316) DS [2.7193305] # 10, 1 SAR
    6 14:28:17 (0018,1316) DS [2.7193305] # 10, 1 SAR
    7 14:31:50 (0018,1316) DS [1.2738155] # 10, 1 SAR
    8 14:34:23 (0018,1316) DS [1.8311454] # 10, 1 SAR

    At 6 months post incident :

    1) Skin – My skin continues to hurt wherever the thermal rays hit it. Face, neck, arms, upper torso and some on upper thighs. Erythema still present on face, neck and upper chest. I have a little temporary relief with lidocaine based topicals and anti inflammatories. Skin continues to atrophy and now shows marked deterioration and scarring. Dermatology consultations refer to skin corrosion consistent with a thermal burn injury and talk about protracted recovery times, and sometimes permanent damage.

    I have documented this injury with many photographs that can be made available if I’m contacted via e-mail.

    I would very much appreciate your feedback on this. What happened to me ? How could it have been prevented ? What can I expect in terms of recovery and long term prognosis. Any recommendations on who else to contact would also be appreciated.


    John Bowler
    Phoenix, AZ

  33. Tobias Gilk Post author


    In the US, orthopedic hardware is frequently (though not always) not a meaningful magnetic risk in MRI. Depending on the particular alloy used, however, hardware may create a significant area of distortion (artifact) that would make diagnosis difficult, if not impossible, for anything in the immediate vicinity of the hardware.

    If the radiologist has given you the ‘green light’, I presume that this decision is based on his/her assessment that there is not any significant risk. There may still yet be image-quality issues, but these aren’t always identifiable prospectively, so it may be a ‘let’s see what kind of image we can get’ approach from the radiologist.

    I hope this helps.

  34. Tobias Gilk Post author


    I’m sorry to hear of your condition. What you describe is substantially different from the acute, localized burns that are typical of MRI/RF burns. While the MRI exam (either directly, or coincidentally) may be related to the onset of symptoms, you may ask your physician(s) to explore the possibility that the skin condition is something that is independent from a possible RF burn.

    I am not a physician, so please don’t take this as medical advice. I’m only identifying what you’ve described as significantly unusual compared with typical reported MRI burns.

    The IEC, which is the international trade group that has set the standards for MRI device safety, has determined that 2 Watts/kg is a level below which there should be no meaningful risk (if the RF energy is not otherwise focused / concentrated within the volume of distribution). Scans of up to 4 Watts/kg of RF energy are fairly commonplace in clinical settings, though obviously with more energy comes more potential for energy-related harms.

    I hope this helps.

  35. Martin

    Hi i want to know what is better the x-ray or magnetic resonance to find metal in body with size of 1 millimeter ?

  36. Tobias Gilk Post author


    When very small, metal fragments can be very difficult to identify on a conventional X-ray, particularly if the foreign body is in an area with lots of bone material. Small pieces of ferromagnetic material may cause (comparatively) large regions of signal loss for some MRI scans. So an X-ray may show it… but it gets lost in other opaque / dense material (bone), and an MRI may not show it (depending on the particular type of scan being run) or an area around it. The best bet for determining the optimum way of finding / identifying a retained foreign body would be to talk with a radiologist.

    I hope this helps.

  37. CJ

    There is a posting from TJ on 6/1/17 that is similar to a concern I have. We also had an MRI trailer moved into our neighborhood. It has taken months for my husband and I to discover that our insomnia is being caused by a low frequency hum coming from the MRI trailer. We actually got up at 3:00am and followed the sound to the MRI trailer. It causes a deep reverberation in your ear drums that is very uncomfortable. Could this possibly be related to maintenance on the machine? Or is it simply that the machine is on all of the time? It is starting to effect our health due to a lack of sleep. Shouldn’t these machines be contained inside a building with soundproof walls? Any suggestions you may have would be very much appreciated.

  38. Tobias Gilk Post author


    I would venture to guess that any sound you’re hearing off-hours (when the MRI is not operational scanning patients) is a result of mechanical equipment to power or cool the MRI scanner or the room, and not specifically from the MRI scanner. In this regard, while it may be a bit bigger than a home air conditioning system of your neighbors, it’s probably fairly similar. You may be able to ask the hospital / clinic / imaging center to provide sound walls to help reduce the noise… particularly if it exceeds local codes or zoning laws.

    I hope this helps.

  39. Mary

    If you occasionally had to go in the mri room to position a patient on the table and you wear a wig with metal clips what could happen? If I wore a headband on top? Would it be ok? Lord knows I don’t wanna take my hair off and be embarrassed in front of staff and patients?

  40. Tobias Gilk Post author


    Without knowing the size or materiality of the clips, it would be hard to say. If you have a strong magnet (something you can hang a calendar on your refrigerator with), you can test the clips you use with that. If there’s a strong attraction to those clips with your magnet from home, you can bet that the attraction from the MRI could be much stronger. That will at least give you the chance to test and perhaps find some less magnetic clips.

    It’s always advisable to share information about anything metal that you have on or in you with the MRI techs. Trust me, they’ve seen / heard more than you could imagine… take them into your confidence and have a conversation with them sometime when you’re not under the time pressures of patient care.

    I hope this helps.

Leave a Reply

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

Change the CAPTCHA codeSpeak the CAPTCHA code