Risky Business

Over the past 30 years, MRI has earned its reputation as one of the safest imaging modalities ever developed. Without ionizing radiation, the concerns with repeated imaging studies, even of children, are largely absent – unlike CT or other X-ray-based imaging. But you can’t have a super-cooled, super-conducting electromagnet that is 30,000 to 60,000 times the strength of the Earth’s own magnetic field without having some safety risks, risks to patients that appear to have multiplied many times over in just the past few years.

Varied reporting structures between states, provider organizations, and the U.S. FDA make it challenging to assemble a complete picture of MRI accident rates. The state of Pennsylvania, as an example, records more MRI accidents and “near-miss” incidents through their Patient Safety Authority than does the FDA, even though the FDA is charged with collecting information from all 50 states.

As flawed as it is, the FDA’s Medwatch program has tracked user reports of accidents and injuries for FDA-approved medical devices across the country for many years. Medwatch data is publicly available via the FDA’s MAUDE database, which can be accessed online through the FDA’s Web site.

Due to gaping holes in the FDA’s accident reporting structure, it is estimated that the FDA collects less than 10 percent of the actual number of incidents. Some experts, including John Gosbee, MD, former co-director of the VA’s National Center for Patient Safety, estimate that the FDA captures 1 percent, perhaps less, of the actual number of MRI accidents. While the proportion of reported accidents remains small, the recent trend that it illuminates is alarming.

Since 2004 (the last year in which the FDA recorded a decline in the number of MRI accidents), the rate of reported accidents has grown almost 300 percent to 2008. This means that a person receiving an MRI today has nearly a four-times-greater risk of an adverse event than they would have in 2004. Why?


The Search for a Culprit

In this four-year time period, MRI utilization has grown at only about 3 percent per year, so only about 13 percent “natural” growth in accident rates might be expected as a direct result of increases in the number of MRI exams. Increased reporting, another potential explanation for the dramatic growth, is not thought to be a factor as there has not been a single governmental or professional body working to increase awareness of reporting criteria. The most logical explanation for the increase in reported accidents is that we are actually harming more MRI patients.

The FDA data is too incomplete to implicate any single risk factor. Additionally, it is believed that a single cause wouldn’t be likely even if the Medwatch data were more comprehensive. Multiple changes in MR technology, clinical application, and operations have each produced sources of greater hazard. These subtle risk-factor changes may be working together, compounding, to increase the total risks to patients and caregivers. These likely include:

• Increasing use of MR image-guided biopsy
• Burgeoning use of MR for emergent/trauma care
• Demographic shifts in patient co-morbidities
• Staff cuts in response to reduced reimbursement
• Growth in clinical applications for higher-acuity patients
• More powerful magnetic fields and time-varying gradients
• Greater attractive effects from improved active-shielding

Any one risk factor, alone, may verge on statistical insignificance, but rarely do these risk factors act alone, particularly in the in-patient setting.

Compounded risk factors multiply the risk for any single patient. And with an estimated 28 million MRI exams performed in the U.S. alone last year, even modest increases in risk produce significant numbers of adverse outcomes. While not the only source of MRI injury, take projectile accidents as an example.

These events, resulting from the introduction of ferromagnetic materials into the MRI scanner room, are likely the most frequent accident in the MRI environment. If each MRI’s annual chance of a projectile accident grew by only 0.1 percent, we would expect to see approximately 28,000 additional projectile accidents per year! (A 2005 ECRI Institute poll of its members found that MRI projectile incidents were greater than all other types of MRI accidents, combined.)

Later this year, the most infamous MRI accident (in which a 6-year-old boy died from a flying portable oxygen cylinder that was brought into the MRI room while he was in the bore) is due to have its day in court. The family of the boy has sued the hospital where the accident occurred, the contractor that ran the MRI for the hospital, the contractor’s medical director, the two MR technologists on duty at the time, and the MRI manufacturer.

The outcome of this single civil trial may do more to clarify and codify the responsibilities of manufacturers and providers of MRI than anything to date. In 2001, when the accident occurred, MRI had escaped all point-of-care safety regulation by virtue of the fact that it does not use ionizing radiation. Despite this fatal accident’s prominent illustration of one of the major risks of MRI, MRI is largely in the same position today.


The Road to MRI Safety

In the intervening eight years since this watershed accident, there has not been a single regulatory or accreditation requirement for MRI safety at the provider level. This is despite three MR safety “best practice” publications from the American College of Radiology and the Joint Commission’s Sentinel Event Alert #38 (Sentinel Event Alerts are the Joint Commission’s strongest patient safety warning). But is this about to change?

The ACR’s MR Accreditation Committee has accepted, in principle, the idea of incorporating site safety elements in the accreditation requirements. What form these are to take is yet to be seen, but it is reported that included safety requirements would be subject to surveyor verification.

And while the Joint Commission may have abstained from providing any surveyor support and allowed their MRI Sentinel Event Alert to atrophy to “paper tiger” status in 2009, next year will see changes. The referenced standard for healthcare facilities buildings, “Guidelines for Design and Construction of Health Care Facilities,” will be updated in 2010 and will dramatically expand the MRI safety standards for the bricks and mortar of Joint Commission-accredited providers. The guidelines document also serves as the standard for 42 of the 50 U.S. states’ departments of health.

Independent of accreditation and regulatory structures, many MRI providers are more frequently recognizing that MRI safety simply makes good business sense. In 2008, the VA’s National Center for Patient Safety calculated that the average cost of a single MRI projectile accident in the VA system resulted in direct costs of more than $43,000. Clearly, the reduction of just this one type of accident could have significant revenue impact.

Because the VA does not operate on a per-procedure reimbursement rate, the $43,000 price tag does not include lost technical or professional revenue, or legal expenses associated with accident litigation. As a result, the median costs per event are likely much higher for providers operating on a per-procedure reimbursement structure, though it doesn’t take a catastrophic accident to take money from your wallet.

Particularly with reductions in reimbursement, patient throughput has become the life-blood of a MRI provider’s fiscal health. Even the accumulated minor ferromagnetic/projectile incidents (such as the cell phone, nail clipper, etc.) add up to hours of lost imaging time and many thousands of lost dollars in potential revenue. Strategies for intercepting these productivity-killers can have the collateral benefit of helping to reduce the likelihood of the headline-grabbing accidents, too.

Projectile incidents make up only one facet of the overall MRI risk profile. And just as there are tools, protocols, and strategies for addressing this risk, so too are there means of reducing risks associated with patient burns, hearing damage, and all of the distinctive MRI risks to patients, caregivers, and equipment.

While it appears that we find ourselves on a very slippery slope of MRI accident rates, the simultaneously good-and-bad news is that the majority of MRI accidents are attributable to “pilot error.”

MRI conditions have shifted under our feet, and this affects our MR staffers most. Verifying the safety of more implants, devices, and conditional criteria have been added to the technologists’ everyday duties. Couple this with the fact that we are now presented with patients with increasingly complex management issues, plus the sheer power of contemporary magnets are making yesterday’s “practice suggestion” into today’s “mandatory precaution.” Simply put, we need more safety attention from our MR providers today than in years past.

The good news is that the tools for managing these risks are already at our disposal. Existing best-practice documents, recommendations for staff qualifications and training, as well as new tools for screening of patients, visitors, and equipment, are all available. It is simply a question of putting these existing resources into practice.

If we ignore the alarming trend of increasing MRI accident rates, we risk finding ourselves, again, in a situation where we have to explain how it is that another MRI patient could have died from a situation that could have readily been prevented by embracing current best practices.


– Tobias Gilk is the president and MRI safety director for Leucadia, Calif.-based Mednovus Inc., manufacturer of the SAFESCAN® family of ferromagnetic detectors (www.mednovus.com). He has also served on the ACR MR Safety Committee and the IMRSER Corporate Advisory Board. He has contributed to the ACR Guidance Document for Safe MR Practices: 2007, the VA MRI Design Guide, and numerous other MR safety resources and publications. Direct questions and comments to editorial@rt-image.com.