Is MRI Any Safer Than CT? Part I: Systems and Safety Considerations


by Richard C.Semelka, Medscape, 12 Oct 2006

The recent release of a warning by the US Food and Drug Administration (FDA) related to the safety (or lack of safety) of high doses of gadolinium chelates being administered in patients with renal failure is the impetus for this month's 2-part Semelka Spin column, which focuses on safety considerations in magnetic resonance imaging (MRI).

Patient safety and service should be a major focus of all radiologists. In earlier writings I outlined the basic tenets of "diagnostic accuracy" and "safety" as the measures by which we should guide our imaging choices. In those Medscape articles I discussed patient safety considerations related to computed tomographic (CT) scanning, with some mention of iodinated contrast used in CT studies. Now, at the risk of alienating the remaining 15% of the radiology community, it is time to discuss MRI safety.

The basic components of MRI safety are machine-related and contrast media-related. The first installment of this 2-part Spin will highlight MRI machine-related safety considerations.

Safety of MRI Systems

The MRI machine is a powerful magnet that exerts a strong force on iron-containing structures within and outside the body (even somewhat distantly outside the body), on battery-operated devices (as measured by specific absorption rates [SAR]), and on local currents in coil devices, other electrical cable structures, and excitable tissues such as nerves and cardiac muscle. It is important to understand that the effect of the powerful magnet is typically only exhibited on iron-containing objects, as most other metals (titanium, aluminum, copper, etc.) will not experience the ferromagnetic forces. Exceptions to this rule are cobalt and nickel, which are also ferromagnetic but trigger a lower effect in a magnetic field than iron. It is also important to understand that larger iron-containing objects will experience greater forces and can therefore be more dangerous than small objects when placed within the area of the magnetic field that surrounds the MRI system. There have been many cases of metal objects being drawn into the bore of the magnet because they were close enough to experience the magnetic field.

Every patient must undergo a rigorous screening evaluation before having an MRI study; during this screening interview, patients are asked whether they have any battery-operated devices or iron-containing metal objects within their bodies.[1] Since the beginnings of clinical use of MRI as a diagnostic tool, patient information has been recognized as a very important component of MRI study, from a safety perspective. I will focus on the most important safety considerations as a reflection of how common this issue is and/or how potentially lethal MRI can be, and will direct the interested reader to more in-depth sources.

The most dramatic contraindication to undergoing an MRI study is the presence of iron-containing clips on vessels within the brain. Probably the greatest number of deaths from MRI studies has resulted from these clips being pulled off vessels and the patient bleeding to death.[2] It was largely as an acknowledgment of how important a problem this can be with MRI that titanium was developed over the past 15 years as a viable surgical metal. At present, almost all neurosurgical clips are made of titanium, and this most serious complication of MRI almost never occurs today.[3] Most deaths related to iron-containing neurosurgical vascular clips occurred in the 1980s, early in the history of MRI.

Check for Implanted Devices

The next most important contraindication to MRI is the presence of a cardiac pacemaker. The powerful magnet can stop all implanted battery-operated devices (eg, cardiac pacemakers, bladder pacemakers, nerve-stimulating devices of various types for inner ear and spinal diseases); however, the cardiac pacemaker is the device that is most likely to result in patient death if its function is interrupted. In the event of accidental imaging of a patient who has a cardiac pacemaker, the first step is to check the patient's pulse to determine whether it corresponds to the rate that the pacemaker is preset to. If there is uncertainty, it is most appropriate to consult the cardiology service responsible for pacemakers. At the present time, the cardiac pacemaker is considered an absolute contraindication to MRI. It should be noted, however, that some pacemakers can tolerate exposure to magnetic fields without interruption of their function, and pacemakers are in development (currently in use in Europe) that are compatible with a magnetic field. Inadvertent scanning of patients with cardiac pacemakers shows that most patients do not suffer injury from this mistake.[4]

Metal shavings in the eyes are an interesting topic to me, as it shows how anecdotal experience can influence medical practice at great expense despite limited data - but this is beyond the scope of this presentation, so I will leave it at this: All patients are screened for possible metal shavings that have penetrated into the eye, and may undergo orbital x-rays to screen for metal.[5]

All metal that may be present in the patient is carefully considered on a case-by-case basis to see whether it poses a hazard, and this is also true for all medical devices that are battery-operated. In cases in which patient safety may be uncertain, MRI facility workers can consult a Web site developed by Frank Shellock, MD, PhD (, to determine whether the metal device is hazardous to patients when it is subjected to the magnetic field.[6]

I am actually amazed that metal does not pose more of a problem for MRI than it does. In most areas of the body (except within the brain), the body heals by scarring. This scarring can hold in place most metal (eg, cardiovascular clips, metal sutures, bowel sutures) after 6 months of it being placed. Performing MRI earlier than 6 months of the metal being placed requires a judgment call as to whether the potential benefit to the patient (over another imaging modality such as ultrasound or CT) outweighs the potential risk. Some metal is so well-seated that it almost never moves on MRI (eg, hip replacements), and some metal is not ferromagnetic and never moves on MRI (eg, tooth fillings).

Magnetic Field Safety: Outside the Magnet

Problems with metal objects within the area of the magnetic field of the MRI system can be the most dramatic misadventures in MRI — with reported cases of wheelchairs, stretchers, hammers, and even old-style oxygen tanks flying to and getting stuck to the bore of the magnet. Perhaps the most famous, and extremely tragic, case in recent times occurred in New York State in 2001 (see, in which an iron-containing oxygen tank flew through the air into the bore of the magnet and crushed and killed the 6-year-old child being scanned. In that instance, a metal oxygen tank was mistakenly brought into close proximity to the patient by non-MRI medical personnel.[7] On the basis of that one case, all oxygen tanks in the United States have subsequently been made out of nonferrous metal (aluminum).

The importance of critical attention to all devices that are brought into the scanning room of MRI is one of the reasons that I greatly prefer CT in the setting of the emergency room major trauma patient. It is sometimes impossible to obtain good patient information in the emergency room, and it is always difficult to pay attention to everyone and everything who may enter the imaging suite when one is imaging a subject who is critically injured and unstable during an MRI scan; whereas with CT there is no need to be concerned with objects brought into the scanning room since they will have no effect on the patient - for example, there is no concern of metal flying through the air with CT.

Another injury that can be sustained during MRI relates to electricity-related injuries that occur when cables are improperly positioned, resulting in a loop configuration that can generate heat and cause patient burns. Attention to all cables and proper cable positioning is critically important to avoid the potential for burns.

The use of MRI in a wide range of patients has resulted in a greater number of unstable sedated patients being scanned. In this regard, monitoring devices for patients' vital signs must be MRI compatible. Careful attention should be paid to life support systems to ensure patient safety while under sedation. Because of the longer tunnel on MRI, heightened attention is needed to visually monitor patients.

Specific absorption rate, or SAR, refers to heat deposition in the body, which is determined by radiofrequency (RF) energy and gradient switching and is measured in units of watts per kilogram (W/kg). There are restraining regulations governing SAR deposition in the body. Current FDA guidance limits SAR whole body exposure to a range of 1.5 to 4.0 W/kg, depending on the patient's clinical condition.[8] SAR limit restrictions are incorporated in all MRI systems.

And finally, another biological effect that can occur with MRI is the stimulation of electrically excitable tissues such as nerves and cardiac muscles. I call this a biological effect rather than a danger because there is no known incidence of injury or fatality through this mechanism. Even so, the stimulation of peripheral nerves can be disturbing and is also avoided per FDA guidelines by restricting how the examination can be performed.

The FDA has extended nonsignificant risk status for clinical fields up to 8T.[9] However, there are some concerns about the effects of high-field strength magnets on humans. The safety issues for such high-field strengths will need further clarification and constant reevaluation. We must always be vigilant in looking at the health aspects of ultra-high-field MRI as we move forward to more and more ultra-high-field systems, and higher and higher field strengths.