Imaging Safety: MRI Safety Today, Six Years Later


November / December 2007

Imaging Safety

MRI Safety Today, Six Years Later

In July 2001, a 6-year-old boy died in a tragic MRI accident at the Westchester Medical Center in New York state. This accident immediately became grist for newspapers and national news telecasts. It also spurred a number of peer-reviewed publications on MRI accidents that hinted at a much broader problem than the “one fluke incident,” which was the popular rationale for deflecting concerns for the industry.

Edward A. Stolzenberg, then president and CEO for Westchester Medical Center, bravely stepped forward in the hours immediately following the accident, accepted the hospital’s role in the accident, and promised transparency in the investigation to help assure that the root causes would be uncovered to help make certain that this sort of accident was not repeated.

Six years after the accident (and only slightly less time since the Mr. Stolzenberg was relieved of his duties for Westchester Medical Center), as the $10 million lawsuit against the hospital, medical staffers, and the MRI equipment manufacturer is poised to start trial, what is the state of MRI safety?

Given the prominence of this accident, certainly it must have already effected significant change in the industry, right? Unfortunately, nearly every current regulatory and accreditation standard for MRI indicates otherwise.

In the immediate aftermath of the tragedy at Westchester — before he was ousted — Mr. Stolzenberg kept his promise. a report developed internally, which identified both causes and recommended corrective actions, was briefly made available. More recent attempts to explore the specifics on the details of the accident have been frustrated by the fact that the original report appears to have been sequestered, perhaps in response to the current litigation.

As a consequence, information on the root causes is not readily available to be shared with the industry, which has been left to make what it can of anecdotal crumbs. Even the public documents made available through requests under New York State’s equivalent of the Freedom of Information Act fail to tell the whole story as there was virtually no documentation of the accident site made by the state investigators.

What is apparent from the various remnants of information available today is that this accident was the result of significant holes in many layers of protection, including operational protocols, staff and clinician training, physical design of the facility, and preventive maintenance regimes. Despite the rich field of areas of potential improvement, the early promise of transparency and analysis to better the industry was only briefly fulfilled, before the current “information blackout” was instituted.

The Authorities
When it comes to issues of medical device safety, the organization many would consider at the “front line” is the U.S. Food and Drug Administration (FDA). However, when it comes to medical devices the FDA’s responsibility ends with the safe “as intended” offers no oversight or authority to operational or facility safety provisions for a sited device. This is not to say that the FDA has not supported MRI safety initiatives, rather that there is not even the threat of regulatory oversight from this agency.

Accrediting bodies, from the broader coverage of the Joint Commission to the radiology-specific accreditation offered by the American College of Radiology (ACR), have not as yet put forward substantive safety requirements as a part of their programs that verify the quality MRI providers.

The single most common (and often the only) Joint Commission surveyor issue remarked upon in the MRI suite is the presence or absence of a non-magnetic fire extinguisher. None of the other factors from the industry-shaking accident at the Westchester Medical Center are survey criteria.

The absence of MRI safety requirements seems particularly incongruous for the ACR, which has now published three editions of MRI safety guidance (published as the White Paper on MR Safety in 2002, updated in 2004, and rewritten and republished in 2007 under the new title Guidance Document for Safe MR Practices;
). ACR accreditation for MRI providers has historically focused on technologist credentials and image quality issues, and has not, to date, incorporated any of the ACR’s own safety charges.

Building codes and standards governing the design of hospitals and outpatient providers have similarly failed to show any changes that address the MRI suite design factors that appear to have contributed to the Westchester tragedy.

In short, the regulatory environment under which MRI providers operate today is virtually identical, in terms of safety, to what it was prior to the young boy’s death.

The Momentum
What little qualitative data exists is beginning to hint at increases in the contemporary MRI accident rate. The FDA’s device accident database (Medical and User Facility Device Experience, or MAUDE) recorded a 143% single year increase in MRI accident reports for the 12-month period ending mid-2006. Many MRI safety experts believe that compounding risk factors, such as greater numbers of patients with implants and assistive devices, increasing use of MRI for interventional procedures, higher magnetic field strengths, and a shortage of credentialed technologists are all working to push up accident rates.

Though, with the paucity of data, it’s not possible to completely rule-out increases in accident reporting as the source of this surge in accident reports.

To illustrate the value of the FDA’s data, one need only compare the accident reports from the state of Pennsylvania, which recently implemented mandatory reporting of accidents and near-misses, to begin to see what the safety experts have been saying for years, namely that the FDA’s data on MRI accidents isn’t worth the paper it’s printed on.

In a 16-month period (from June 2004 to September 2005), the Pennsylvania Patient Safety Authority logged 88 MRI incidents. If you extrapolate that out over 10 years, that’s 660 incidents. The FDA’s database of MRI accidents throughout the US for a 10-year period ending in 2006 logged only 389 MRI incident reports, or slightly less than 60% of the presumed 10-year rate for the state of Pennsylvania alone!

It’s no surprise that Dr. John Gosbee, former co-director of the VA’s National Center for Patient Safety, speculated that the FDA’s MAUDE database records 1%, perhaps less, of the actual number of MRI incidents.

The Changes
Though the regulatory environment relative to MRI safety has yet to see any significant change in the wake of the Westchester tragedy, that truth may be short-lived.

A. Joseph Borelli, MD, chair of the ACR’s Committee on MR Accreditation, has indicated that his committee will begin reviewing the current ACR Guidance Document for Safe MR Practices: 2007 with an eye towards incorporating safety standards into the ACR’s accreditation process. Particularly given the increasing use of ACR accreditation as a payer credentialing process, this one step may apply MRI safety requirements to thousands of MRI providers.

The Joint Commission has purportedly been working on a MRI safety Sentinel Event Alert (SEA), their highest cautionary alert for patient safety, since November of 2006. a recent teleconference hosted by the Joint Commission indicated that this was the next scheduled SEA and was due to be released shortly. One can only assume that this will be followed by new MRI safety standards and surveyor review criteria.

The Guidelines for Design and Construction Health Care Facilities, the most predominately cited building code for hospitals and healthcare providers, is currently in the midst of a revision cycle for a 2010 republication date. The Guidelines authors are currently collecting and reviewing proposed changes to the code, which include substantive revisions to the facility design criteria for MRI providers.

Individual MRI equipment vendors, healthcare systems, and many lone providers have taken it upon themselves to proactively respond to the available MRI safety “best practices,” though implementation is inconsistent across any broad measure.

While many may have expected decisive standards for the protection of MRI patients and staff shortly following the 2001 fatal accident at Westchester Medical Center, that was not the case. Many of the critical facts have been obscured, presumably in response to the ongoing litigation. Many organizations charged with the protection of MRI patients have been slow to implement policies to mitigate those risks that have been known for years. However, this dangerous trend towards inactivity appears to be on the verge of change as many of the authorities that can effect significant change are poised to do so. Hopefully each will overcome the last 6 years of inactivity, deploying standards in time to prevent the injuries or fatalities that are virtually guaranteed in their absence.

Tobias Gilk is an imaging design consultant with MRI-Planning, an imaging facility consulting firm in Kansas City, Missouri, that specializes in MRI and multi-modal facility planning, design, and construction. Gilk also serves on the American College of Radiology’s MR Safety Committee, which recently released the ACR Guidance Document for Safe MR Practices: 2007. He can be reached at