Electromagnetic interference can cause hospital devices to malfunction, McGill group warns

Canadian Medical Association Journal 1996; 154: 373-375

Top of document

Introduction

This is an example of a growing problem -- electromagnetic interference (EMI) from sources such as television transmitters, police radios and cellular phones. EMI can cause monitors and other medical devices to malfunction, says Bernard Segal, principal investigator of the McGill University Biomedical Engineering Group on Electromagnetic Compatibility. It was created in 1989 to study, predict and prevent such malfunctions.

The group also hopes to educate users, including physicians, nurses and patients, on how to spot such problems. "Medical professionals must play a crucial role in identifying malfunctions caused by electromagnetic interference," says Segal. "They are on the front line. If they don't see a medical device malfunctioning, the engineers and researchers can't do a thing to solve the problem."

EMI occurs when electromagnetic waves emitted by one device interfere with the normal operation of another. Sensitive electronic components inside any equipment that relies on computer chips can be vulnerable to other electromagnetic energy. Electromagnetic compatibility (EMC) refers to the capability of electronic devices to function properly in an electromagnetic environment.

Segal emphasizes that EMI events are rare, since they occur only under certain conditions. Although a medical device may or may not be susceptible to EMI generated at a particular frequency, it may be sensitive to many widely separated frequencies simultaneously. A weak source has to be close to the device to cause interference; however, a powerful source may cause EMI even if it is kilometres away.

EMC problems have been recognized and dealt with for years in the telecommunications, aviation, automobile and computer industries, but today biomedical engineers are also familiar with them. In the health care field, however, physicians and nurses are not trained to be aware of EMC problems. Furthermore, these problems can be particularly hard to identify and prove.

"They are often transitory," says Segal. "A malfunction might be due to a security guard walking down the hall with a walkie-talkie, or an ambulance outside the building. And staff don't have the time or resources to track down the source."

The result of medical-device malfunction ranges from death or serious morbidity to mere inconvenience. Efficiency also is reduced if a device isn't working or must be reset. Other incidents are merely nuisances -- when alarms go off, for example.

These problems are emerging in health care because much of medical equipment that was built 5 to 10 years ago, when EMI was not recognized as a serious issue. Since then, says Segal, "there has been an explosion of technology, and an explosion of information usage."

In future, hospitals will have complex wireless communication networks to deliver patient information wherever it is needed. Although this will have a positive effect on health care delivery, it will also lead to more EMI problems because many medical devices were not built to function in this kind of electromagnetic environment.

Home care medical devices are also vulnerable. "[Physicians] are going to send people home with medical equipment to an even more complex electromagnetic environment," warns Segal. "You have no idea of the quality of power in the building. There may be a ham-radio operator next door, or a police car outside. And if a medical device malfunctions in the home, there may not be someone to notice the patient doesn't look right, or that the machine is doing something it isn't supposed to do. Dealing with electromagnetic compatibility is a potential life-and-death problem."

Segal, who feels this is the most important project he has ever undertaken as a research scientist, has a background in physics, biomedical engineering and neurophysiology. He is research director of otolaryngology at Montreal's Jewish General Hospital.

The McGill group includes electrical engineers and the heads of biomedical engineering at five university-affiliated hospitals. There is also a legal expert who deals with questions related to liability, and a consultant who specializes in the effects of radio waves on biologic systems.

Researchers face two challenges, says Tomás Pavlásek, emeritus professor of electrical engineering at McGill: the technical puzzle of identifying and minimizing incidents of EMI, and human nature. Instead of denying responsibility for the problem, all parties -- those who manufacture and maintain medical devices, those responsible for the sources of EMI, medical-device users, government, researchers and those who set standards and regulations -- must collaborate to find a solution.

Much progress has been made in the last year, Segal acknowledges. The McGill project received funding last year from Bell Mobility and Cantel. In the United States, the wireless communications industry has sponsored the Center for the Study of Wireless Electromagnetic Compatibility at the University of Oklahoma. In 1994, Health Canada's Medical Devices Bureau hosted a roundtable discussion in Ottawa for groups interested in EMI, and a cross-border Task Force on Electromagnetic Compatibility in Health Care has been established, with Segal acting as coordinator.

Top of document

EMI realities

Devices that are used to measure patients' health indicators are particularly susceptible to interference for two reasons: they can measure very low-level signals, and the human body and the connecting wires act as an antennae.

The malfunctioning neonatal radiant heater cited earlier was an example of one of the most difficult EMI problems -- a "silent" malfunction, in which the device malfunctioned slightly but did not always sound an alarm. Morbidity related to the malfunction was unlikely, but babies may have been slightly overheated or underheated.

There have been more serious problems. US data indicate that some sleep-apnea monitors failed to sound an alarm when babies stopped breathing. Investigators from the Center for Devices and Radiological Health (CDRH) of the Food and Drug Administration found monitors erroneously detected respiration when exposed to low-field EMI. Power wheelchairs started rolling after their brakes released because of certain field strengths, and anesthetic gas monitors stopped working when influenced by interference from electrosurgery units.

CDRH reports include an external defibrillator/pacemaker that stopped working when an ambulance attendant used a hand-held transmitter too close to a patient, who was not resuscitated. As well, an infusion pump's rate changed when a cellular phone was placed on the instrument stand, and interference from a computed-tomography system caused a problem in an ultrasound scan, resulting in a misdiagnosis.

Some malfunctions remain mysteries. "In one Quebec hospital, a malfunction happens at a particular time each week," says Segal. "A bunch of monitors don't work. It's just a nuisance, but the hospital doesn't have the resources to track down the source."

Many new devices are designed to be more resistant to unwanted electromagnetic signals -- in effect, the equipment is "immunized." However, Pavlásek says immunization is "a major technologic problem given that [it] involves EMI sources operating over an extremely broad spectrum, and given the exceptional electromagnetic complexity of a patient in an [intensive care unit] when connected to a range of equipment. For example, try counting the number of `antennae' [connecting wires] that are connected to a typical patient, noting that even fluid-filled infusion lines constitute antennae. . . ."

Standards will also help. The European Community recently set new immunity standards for electromagnetic compatibility for medical devices sold in Europe, and in 1993 the International Electrotechnical Commission published some standards for EMC. However, standards don't offer a complete solution because they only apply to new devices. Much of the equipment currently in use has never been tested for EMC, and testing is expensive.

Furthermore, it would be impractical to test every medical device under simulated conditions for every possible rare event and set of operating conditions. As new technologies are developed, new sources of EMI will arise. And as soon as a device is connected to a patient, standards do not apply.

Participants at the recent Health Canada medical devices roundtable reached several conclusions. They agreed that a total ban of radio-frequency transmitters in hospitals was not justified, and suggested suspectable equipment should be removed if the source of EMI cannot be controlled. They also recommended rational management of wireless telecommunication devices used in hospitals.

Segal explains that steps to managing the electromagnetic environment in hospitals may include increasing the distance between sources of EMI and susceptible devices, lowering the power of EMI sources when possible, removing or relocating devices that are highly susceptible to EMI, and labelling susceptible equipment. Staff and patients also should be made more aware of EMI problems. Administrators should consider electromagnetic comparability when purchasing new medical equipment, even though the cost may be greater. Hospital planners should also consider EMC, since most walls and floors are not designed to shield equipment from electromagnetic energy.

Although some hospitals have already banned cellular phones and wireless devices, Segal suggests this may be unnecessary: "To my mind all walkie-talkies, cellular phones and wireless devices should be excluded from critical-care areas, unless it has been demonstrated that a particular wireless device is safe in that environment. Then you should designate areas where [EMI] sources such as cellular phones can be used."

As medical devices and wireless information networks become an increasingly essential part of health care delivery, problems with EMC will continue to grow. "These are problems about tomorrow's health care system that have to be dealt with today," says Segal.

Tips for spotting EMI problems

• When a device behaves inexplicably in a patient area, but works properly in the biomedical
• engineering facility, or in another location, EMI may be the problem. The patient area may be exposed to a source of EMI, but the repair shop, which is often in a basement, may not be. Thus, the problem may not be apparent there.
• EMI may be the reason if a patient's condition does not appear to match data provided by a medical device or monitor. Do not place absolute trust in technology.
• Malfunctions may occur intermittently when other devices or telecommunications equipment are used in adjacent areas, or during certain weather conditions.
• Age may be the culprit if a device that has always functioned normally begins to develop problems.
• If an alarm goes off inexplicably, someone may have passed by with a walkie-talkie or some other device.

For more information, contact Dr. Bernard Segal at the Department of Otolaryngology, Sir Mortimer B. Davis - Jewish General Hospital, Room E-266, 3755 Côte Ste-Catherine, Montréal PQ H3T 1E2; 514 340-8222, x5987; 514 340-7581 (fax); Segal@Jghvax.Medcor.McGill.Ca (email).