| |||||||||||||||
|
Home
|
For other articles and previous issues click here. November 28, 2005 How
Not to Design an MRI Suite — Part 3 in a Series on Facility
Safety Many MRI facilities seem to continue operating on the age-old principle that what you don’t know won’t hurt you—until, of course, it does. Four years after the much-publicized Colombini incident, the floor plans of many typical MRI suites put both patients and staff at physical risk—not to mention sizably increasing the facility’s liability burden. That’s according to medical architect Robert Junk, who specializes in designing healthcare facilities and frequently speaks on the subject of MRI design and safety. In his view, neither hospital administrators nor building designers have done enough to apply the real lessons learned in the aftermath of that tragedy. “We find clients aren’t necessarily aware that their architects or engineers aren’t giving them the latest safety features in their designs,” says Junk. To combat this attitude and help shed light on too-long-neglected MRI safety issues, Junk uses full architectural renderings in his seminar, “How Not to Design an MRI Suite,” most recently presented to the American Healthcare Radiology Administrators (AHRA) and at the National Symposium on Healthcare Design, among others. Reviewing the Evidence In reviewing the hospital and police accounts of the accident, Junk and his colleague Tobias Gilk believe that the design of the MRI facility at Westchester Medical Center at that time presented significant physical hazards. “There are a number of architectural design interventions that could have broken the chain of events,” says Gilk. “Probably any one of them, and certainly in combination all of them, could have kept this accident from happening, or at least interrupted it at any given point.” Figure 1 (on facing page) recreates the 2001 floor plan and shows the sequence of events. (The diagram here is an approximation based on public records.) Tracing the actions of the staff through the suite rooms, it’s easy to see how physical features may have helped prevent the accident: 1. The oxygen system wasn’t linked to the rest of the facility. “One of the main issues of this [suite] being a kind of appendage on the hospital was that the oxygen supply for the scan room was not part of the main medical gas system,” Junk explains. “The MRI unit had its own independent system and therefore was not alarmed to alert anybody inside the main hospital when it failed.” The technologists on duty had to leave the control room to access the external oxygen supply. 2. There was no quarantine area where non–MRI-safe objects could be securely locked up. Says Gilk, “The Westchester technologists did exactly what they should have: They identified something that was unsafe for the patient [the oxygen cylinder] and removed the object. But they didn’t follow through with the next step, which is to lock that item up so that it could not inadvertently be brought into the scan room. In such a tiny modular building, it can be difficult to find the necessary space, although all you really need is a storage cabinet that’s lockable. [A cabinet] could have been placed on any one of these walls, inside the control room, even in the vestibule area directly outside the toilet, as long as the technicians on duty are able to secure it.” 3. The suite was too easily accessible to nonauthorized persons. “The scan room door should have been locked, or at least there should have been some sort of security features to prevent people from entering at will,” notes Junk. For example, because there is no dedicated MRI patient holding room or secured access, the nurse became involved while simply passing through a public area that impinged on the MRI space. Clearly the accident was a case of human error, but bad design tragically compounded the results of that error. Yet the Westchester facility represents a fairly common layout plan, according to Gilk. “We’ve seen these issues in multiple cases, where MR scanners and MRI suites are shoehorned into spaces that are really a fraction of what they ought to be [for safety and efficiency], without allowing necessary space for screening, quarantine, storage for MR-safe items, all that. None of these problems are unique to the Westchester design.” Not-So-Safe Assumptions “MRI equipment vendors provide design professionals with CAD templates that appear to be detailed down to the Nth degree, showing exactly what you need for the suite,” notes Junk. “To the untrained eye, they look complete, and they look official. So you assume this must be everything you need to know. “The problem is, vendor templates are great, useful pieces of information for telling you how the magnet is going to operate, but that’s all… Vendors have made it very clear that the only thing the template tells you is the physical requirements to make the magnet operate—how big the room needs to be, where the control room needs to be, the size of the equipment room, [and so on],” he adds. “Unfortunately, many people fail to [discern] where the vendor’s responsibility ends and the facility’s begins.” The results of that tradition are exemplified in Figure 2 (on page 38), where, says Gilk, the designers essentially “cut-and-pasted” a vendor-provided template to fit into the space available on the existing floor plan. “For example, the vendor documentation might say, ‘We recommend you provide a transfer grill and an emergency exhaust duct,’ but it’s left to the architects and engineers to decide where is there a convenient place [to] drop it in, without projecting the consequences. The fact that the design of the [Figure 2] example actually created new safety hazards is indicative of the way that many [planners] rely on vendor templates as gospel.” As another example of what he calls “the pervasive false assumption that everything on the vendor template is perfectly complete,” Gilk points out the lack of hand-washing sinks in nearly all MRI templates: “Why are there no sinks in the MRI suite when we’re doing needle-guided biopsies and contrast injections? The only reason they aren’t there is because they don’t show up on the vendor templates, so nobody thinks about [them being missing]. The assumption is, if they’re not there, we don’t need them.” As he points out, such assumptions can have a direct negative impact on patient care and infection control issues. Gilk continues, “Twenty years ago, when the technology was so new and unfamiliar, architects and engineers, facility planners, even technologists, were virtually completely dependent on the vendors helping them work out the details. But we know an awful lot more now about what can and should—and shouldn’t—go into these suites.” A Host of Problems There is no clearly delineated MRI patient screening and holding area. Gilk points out that a curtained-off holding area for CT seems to be a shared MRI/CT patient waiting station. “But there are two doors right in front of the shared area. The patient who has a pacemaker or is on oxygen who gets up to use the bathroom, or to ask a question, and who wanders through the wrong door could make a fatal mistake.” The plan does not designate a quarantine area for nonsafe equipment that can be locked down but still accessible as needed by staff. • Entrances are not identified as lockable. Says Junk, “From the plan, we can’t tell if they have any system of controls to prevent unauthorized people from simply walking in, and right up to the scan room door.” • Too many entrances. The more entrances, the greater the likelihood that someone will bring something into the scan room without notice. At that point, you’ve missed your chance to screen them. In fact, a recent client on recommendation revised a floor plan to allow only one means of entrance to the MRI suite. “And they will position a ferrous-only metal detector at that entrance, so they can both manually and through technology screen patients and staff as they come through this ‘gate’ area,” he says. • The magnet room door swings inward. During a quench, pressure from escaping cryogens could pin the door shut, trapping anyone inside. While the cryogens themselves are not toxic, they displace all the existing air in what is now a sealed chamber, creating the risk of asphyxiation. The gases are also capable of freezing human tissue on contact. “Again,” notes Junk, “the door swings in because that’s how it’s shown on the original vendor-provided template.” As noted earlier, the transfer grill is essentially useless. The transfer grill allows pressure to equalize from one place to another, thus reducing the positive pressure buildup, during a quench. However, in this plan, Gilk explains, “the cut-and-paste method has all the right pieces, but they have been put together in the wrong order. Probably the facility administrators—and perhaps even the architects and engineers—had the illusion that all these important safety features are in place, but the way in which the design was executed means that those features won’t work the way they were intended. In fact, [constructed as shown] it could present new hazards.” • The cryogen exhaust duct is positioned over the exit door. Thus, if a cryogen leak occurs, the automatic exhaust fan will actually draw the escaping super-cold gases directly into the path of anyone trying to get out of the room. The only way to exit is to run through the hazard you’ve been trying to avoid. Notes Junk, “All they had to do was move the exhaust duct to the back of the room.” • The MRI function shares a control room with a CT unit. You need a dedicated path to bring patients into the MRI, contends Junk. “Bringing other patients along that same path just increases your need to do screening. Someone going to the CT doesn’t need the same level of screening. [This design] effectively doubles the work done by staff—or else you’re not screening … and you’re potentially bringing patients with implants or pacemakers or other devices too close to the gauss field of the magnet.” Learn From Others’
Mistakes Incorporating safety features in the suite’s physical layout—including a screening area at the inpatient entrance, special intake and patient holding areas adjacent to but separate from the outpatient entrance, and control rooms with full line of sight all nearby doors—virtually eliminates all the problems illustrated in the two designs shown. Junk and Gilk note that they don’t intend any of the examples shown here to be used as “standard” templates. Rather, they’d like to see administrators, designers, and engineers pay more attention to each facility’s unique requirements, system specifications, and specific quirks to accommodate appropriate safety features. For example, says Gilk, it’s common practice to use prefabricated modular buildings without modeling how needs may change over time. “We are proponents of modular buildings, but we see a number of modular MRI facilities that are 300 square feet or more smaller than they ought to be … [because] by the time you put the scanner in the control area, the equipment room, there’s nothing left over for a quarantine area or a separate patient holding area.” While it’s impossible to create a universal design suitable for every installation, a good architect should be able to create the plan that will be the most forgiving to eventual equipment changes, adds Junk. “If all you do is design and build a space to the bare minimums of a particular piece of equipment, you’re handcuffed with regard to future options, in terms of new modalities, higher field strengths, and many other issues.” See the accompanying sidebar for more details; for additional examples of both good and bad MRI layout, visit www.MRI-Planning.com/AHRA/2005_AHRA_MRI_Safety_Handouts.pdf. Beyond the White Paper “I think a lot of these issues fall between the AIA and the ACR,” says Junk. Without any regulatory agency mandating new standards, Junk believes independent professional organizations like these are reluctant to risk alienating members by trying to force the issue. Of course, few in healthcare want to see more regulation being imposed from outside. To prevent that, Junk thinks it’s up to industry professionals—including technologists, radiologists, and administrators—to educate themselves and expect more from their engineers and designers. If they don’t, and fairly soon, the eventual outcome of ongoing litigation may drive the expectation of de facto standards enforced by insurers and manufacturers. “If manufacturers face liability [for not mitigating or explaining design problems], one of two things will happen,” he predicts. “They’ll either start to address safety in some of their templates, or they’ll stop giving out the templates. And I bet the latter happens—that’s the one that doesn’t cost any money, and doesn’t open them up to any more risk.” — J. K. Bucsko is a freelance healthcare writer and editor based in Westville, N.J.
For example, he says, “installing shielding for the higher-power magnet now is pennies on the dollar, compared to having to rip the whole room out five years from now, or two years from now, and replacing it. The same thing with floor structure: A little more support now probably makes sense, because having to come in and rework an entire structural system to support a heavier, larger system is going to be a major expense at any time.” The payoff for boosting your initial outlay can be dramatic and continuous, according to associate Tobias Gilk. “Anything that can be done to minimize the ferrous content, vibration interference, moving metal interference near the magnet, and other common problems ensures you get the optimal image quality for that particular magnet. What that means over the life of your system is that you’ll be able to coax another year or two of solid clinical value out of this expensive equipment. Because you’ve protected image quality from the dangers inherent in [some] designs, your financial advantage is going to be huge … even when another whiz-bang new system comes out five or more years down the line.” He points to one client who is only now replacing a 15-year-old 1.5T system—a remarkable lifespan for an MRI scanner. Gilk says the hospital has been able to sustain its use for that many years because the suite was specifically designed to protect image quality. “Comparable installations elsewhere pulled their magnets out five, six, maybe at most seven years later. [This hospital] squeezed years of valuable clinical life out of their equipment as a direct consequence of the features incorporated in the original physical design.” Almost as a bonus, they found when siting the new larger 3T system that they were able to reposition various elements to allow tightening the FDA exclusion zone. “We were able to move the control console and shift the wall, and thus contain the gauss field … so we freed up thousands of square feet of [the hospital’s] real estate that had previously been unusable because of the previous scanner’s larger magnetic field,” he says. Gilk recommends considering every possible feature you may eventually want in your suite whenever you need to do any work. “If you’re interrupting patient throughput anyway, think ahead. Even if you’re convinced you don’t need a sink, for instance, at least put the plumbing in now, so when things change you can upgrade in the shortest time possible without interrupting operations again.” The final word: “You’re probably always better served by inflating the manufacturer’s sizing just a little… A little more money on the front end is going to give your MRI system longer life and greater use down the road,” says Junk. “It’s not going to get cheaper to shut the magnet down five years from now.” — JKB
|
 |
|
3801 Schuylkill Rd • Spring City, PA 19475 Publishers of Radiology Today All rights reserved. |