BBI Contributing Editor

MINNEAPOLIS, Minnesota — This year's Association for the Advancement of Medical Instrumentation (AAMI; Arlington, Virginia) meeting drew more than 1,500 biomedical and clinical engineers from across the county to the Minneapolis Convention Center. The gathering was a venue for presentations on the role of hospitals in terrorist or natural disaster scenarios, with two sessions providing some helpful insights. With healthcare depending upon a strong technological base and disasters — or at least the potential for them — becoming a more common experience, the role of biomedical engineers is changing and their contribution is becoming even more essential.

In a session chaired by Jim Keller of ECRI (Plymouth Meeting, Pennsylvania), disasters in Japan and Houston were discussed. The first focus was the Serin gas attacks in Tokyo subways that affected 5,500 people and ultimately killed 12. The other discussion cited the impact of last summer's flooding in Houston that decimated several hospitals. Vendors and hospitals alike can learn from these unfortunate events, things that conventional disaster drills so far have not revealed. Some of these revelations could lead to the redesign of medical devices, to make them more versatile and robust in such emergency situations. While the two events reported at the conference were fundamentally different, some of the issues that emerged were identical. These two events — one natural and one terrorist-caused — suggest there is a new market for disaster management products, services and planning that existing vendors may wish to exploit. The two disasters differed in their duration and the degree to which their technology infrastructures were affected.

In the Japanese terrorist event, the infrastructure of the hospital that received the majority of victims was fully functional and was sustained during the short duration of the event. The major problem was dealing with 640 of the 5,500 people exposed and not knowing what sort of illness, chemical or biological agent was involved. Ultimately there were 12 deaths, but 40% of the hospital's nursing staff caring for patients also was affected by second-hand exposure to the chemical agents involved. This makes it very clear that such events represent a major risk to hospital staff, as well as those exposed initially. The key was very rapid mobilization of resources from outside of the hospital, including dozens of portable beds, temporary phones and communications devices, food for staff, etc. Knowing where such resources exist, and then how to contact other hospitals and sources to obtain them in an emergency, is among the lessons to be learned from this event. Ultimately, decontamination and limiting the exposure of other members of the public, such as bus and taxi drivers and others who came in contact with victims, also is a matter for consideration. Had the agent been biological rather than chemical, the entire population of the city would have become quickly exposed and at risk.

In the case of the flooding in Houston, the technological infrastructures of the hospitals affected essentially collapsed, and the duration of the event was sustained, complicating the care of existing patients and their families. In this way the two events were fundamentally different. However, a terrorist attack using a contagious biological or radioactive weapon, rather than a chemical agent, as in Japan, would compromise the healthcare system in very similar ways to what occurred in Houston, as the effects would be long-lasting and could lead to a compromise of the technological infrastructures of the hospitals involved. There was much secondary biological contamination in the Houston floods, which rendered tens of millions of dollars of medical equipment useless while it was occurring and worthless after the event.

The natural disaster that affected three hospitals in the downtown Houston area was brought on by torrential rains totaling an inch per hour for almost 40 straight hours. The result was massive flooding, covering freeways with water five feet deep and flooding the bottom two levels of two hospitals. What resulted was a loss of all infrastructure at those facilities. Electrical power was lost, and so was emergency generation a short time later. Without power, all life support was lost but so was all heating and cooling, all communications, all lights, elevators, basically everything that was not battery operated. It became impossible to prepare meals, and the hospital had only the light that came through outside windows during the day. It was determined that patients would be evacuated from the roof by helicopters, since the roadways into the hospital were all under water. To get patients to the roof, however, they had to be carried up unlit stairwells, as no elevators were working. To provide visibility for this physically challenging task, some were posted in the stairwells with battery-operated flashlights so staff could see to carry patients on stretchers up the stairs to the roof. Obviously this was a slow and risky procedure for the more critically ill patients and took several days in all to complete.

Meanwhile, simply living in the facility became challenging. Since no food existed, meals had to be located at fast-food facilities near the hospitals. Staff, patients and families were unable to leave and had to be accommodated with sleeping facilities. A critical lesson learned was that any hospital needs to be prepared to operate completely without outside assistance for a period of at least 72 hours, as that is the minimum time it takes the Federal Emergency Management Agency to mobilize in such emergencies. Communications were established using hand-held radios, but the inventory of batteries and many other supplies was kept in the basement areas, which were quickly flooded. The biomedical engineering areas also were on those levels, rendering them inaccessible and thus making maintenance or repair of anything else impossible. Basic tools couldn't even be retrieved.

Coming out of these experiences were lessons that every hospital should learn from. First, at least three days of all critical supplies should be available on-site, and stored on floors above levels that could possibly flood. After $8 million in losses in biomedical engineering alone, that department was relocated upward in the Houston hospitals to ensure that it would never again be lost due to water flooding lower levels. New, watertight, flood doors were installed in lower levels. An inventory of all battery-operated devices should be maintained, as well as an inventory of their batteries and code dates. Batteries need to be moved in and out of this stock, to assure that all batteries stored are fresh when needed and not discharged or out-of-date. Stocks of mobile radios need to be available. A plan for where people go and how they coordinate and communicate with each other and with the outside world needs to be worked out and practiced. Such a disaster needs to be simulated, including the controlled shut-down of as many systems as possible in order to prepare staff to operate effectively in the event of a real emergency. Just knowing what to do and how standard operations should change during such an event can be enormously helpful when the real event occurs.

A number of suppliers of biomedical engineering management software packages were on hand at the AAMI meeting. These packages help biomedical engineers keep track of the equipment they have, its status, when preventive maintenance is required, what parts inventory is on hand, what is on order, etc. Philips Medical (Andover, Massachusetts) was showing its multi-vendor program software products, a palm-based, mobile approach that works wirelessly with a central server. Indeed, many of these products are becoming personal data assistant-based, allowing technicians to interact with these systems wherever and whenever they need to. Some are using conventional PDAs, but many are selecting ruggedized PDAs, such as those supplied by Symbol Technologies (Holtsville, New York). The addition and integration of bar-code scanners for example, allows these same devices to be used to track inventory asset tags on the equipment being handled.

New products

While not the most common venue for new-product introductions, AAMI did see the introduction of a new member of the Nihon Kohden (NK; Tokyo) family of monitors, one that includes some very advanced features. The use of universal connectors with the transducer section incorporated into the sensor cables, allows many different types of sensors to be plugged into any available slot, a nice feature that makes rapid connection and reconfiguration of the NK monitors possible. Until the AAMI gathering, the Nihon Kohden family was robust in lower- and mid-acuity products but lacking a truly higher-acuity bedside monitor. That will be addressed with the introduction of the new BSM Lifescope monitor, currently pending FDA 510(k) clearance later this fall. The new monitor was shown as a work-in-progress at AAMI.

The new Lifescope easily displays six waveforms, with user-selectable numerics on the left or right side of the screen. It features arrhythmia (in the bedside), ST-segment and full disclosure capabilities. The user interface is a combination of touch screen plus a set of dedicated control buttons. These include a home key that quickly takes the user to the top menu level, plus a menu key and two keys to control the built-in non-invasive blood pressure. There also are two thumbwheel knobs completing the devices dedicated controls.

The sensors all connect on the left side of the monitor, augmenting the core set of built-in parameters — ECG, SpO₂ (from Nellcor) and NiBP. There are two dedicated temperature plugs and five additional universal sensor connections, all of which can accommodate any combination of FiO₂, invasive blood pressure, CO, temperature, respiration or CO₂ sensors. This universal approach makes the monitor very flexible in all higher-acuity care settings. Indeed, the sensor cable connector houses the electronics that make this universal connector scheme on the monitor work. There also are network and printer connectors on the side of the monitor, allowing it to communicate with a remote central station, or to drive a PC-compatible local recorder on alarm. The value equation for this monitor will be aggressive, as we estimate the pricing will be in the $15,000 to $16,000 price range.

Nihon Kohden also was showing a slick Fujitsu computer acting as a patient monitor display. This was a new Jupiter-class Pocket PC (soon to be Pocket XP) computer loaded with patient monitoring application software and is certainly a candidate for a mobile central station-type device, since it is small and light enough to be carried around by a caregiver.

This new device is small and light enough to hold in one hand but has a very readable, high-resolution color LCD display, capable of showing tabular displays of more than 36 rows by 10 columns of four-digit alphanumeric values, for example, on one screen. The user interface includes a touch screen augmented by a set of dedicated numeric keys, four directional keys, and several additional browser and special function keys. The screen also has a set of hot-labeled, application specific keys along a simulated bottom row of the display, which also is annotated with the patient's name and current time.

The device was very flexible. It was also shown displaying four patient waveforms, including all vital signs and alarm limits (on the left of the screen), trends (vertically) of ST-Segments in four separate ECG leads simultaneously, a simulated three-trace recording (complete with calibrated recorder paper background grid and color highlighting of ectopic beats), 12 (horizontal) continuous one-minute ECG strips (again with ectopic highlighting) and conventional arrhythmia detail strips with normal and abnormal beats labeled.

These new offerings, when fully integrated into the family of Nihon Kohden products available in the U.S. market, will again provide hospitals with some cost-effective and interesting alternatives to the largest vendors and their higher prices. It is clear that some original thinking has gone into these new NK models. Given the stranglehold that large group purchasing organizations have on many U.S. hospitals, it is another question whether even such advanced products will find an interested market among such institutions.

Nonin (Plymouth, Minnesota) was showing its new Avant 2120 monitor, a combination of noninvasive blood pressure and pulse oximetry. This compact unit is suitable for general ward use and is aggressively priced at $2,200 list. It is the latest member of a growing family of products.

Infusion safety has become a hot issue since the introduction by Alaris Medical (San Diego, California) of its new Medley pumps and Guardrail software, which it had shown earlier at the American Association of Critical-Care Nurses (Aliso Viejo, California) meeting. The Medley system differs from other pumps in that it also includes some ancillary parameters, such as Masimo's (Irvine, California) SET SpO₂. This allows the software to look at more than infusion parameters in accessing infusion safety profiles. Seeking to level the competitive landscape, Baxter (Deerfield, Illinois) was showing its Colleague CX series with new Guardian features that track dose limits for selected drugs. Doses outside of preset safety limits require intentional override by users and are annunciated on the pump and in the history log. With both Baxter and Alaris offering safety software products for their infusion pumps, the onus of responding now shifts to Abbott Laboratories (Abbott Park, Illinois), the other major supplier of inpatient infusion devices.