IJBEM logo
International Journal of Bioelectromagnetism
Vol. 4, No. 2, pp. 169-172, 2002.

previous paper

next paper

www.ijbem.org

ELECTROMAGNETIC INTERFERENCE OF MEDICAL DEVICES
AND IMPLICATIONS FOR PATIENT SAFETY

Howard Bassen
U.S. Food and Drug Administration, Center for Devices and Radiological Health
Rockville, MD  20852 USA 

Abstract: Hundreds of problem reports demonstrate that a wide variety of sources of radiated electromagnetic (EM) fields can cause malfunctions in many types of medical devices. The most important characteristics of EM fields that affect electromagnetic interference (EMI) are (1) field strength (2) frequency (3) modulation and (4) the voltage induced in the leads and electronics of a medical device.  Some of the most significant sources of potential EMI are wireless (cellular) phones, hand-held transmitter/receivers (such as Citizens Band (CB) radios and guard radios), electrosurgery devices, diathermy, magnetic resonance imaging (MRI), transmitter towers for radio and television, and anti-theft devices.  Medical devices can be made less susceptible to EMI by incorporating special features during the initial design of the device.  Shielding the electronics so that they are completely encased in an electrically-conductive enclosure, and filtering input and output wires can make medical devices highly immune to EMI.  Another approach to control EMI in medical devices is to issue warnings, recommended practices, or contraindications. In perspective the number of EMI incidents is very small, even considering the vast amount of under-reporting that exists.  However, the EMI problem needs to be addressed as the world becomes more “wireless”. 

INTRODUCTION 

This paper presents an overview of the issue of radiated electromagnetic interference (EMI) of medical devices starting with a tutorial for clinical and biomedical personnel.  Conducted EMI, another important problem where interference travels on wires from the EMI source to the “victim” device, will not be discussed here. The U.S. Food and Drug Administration and many other organizations worldwide have received hundreds of problem reports that demonstrate that a wide variety of sources of radiated electromagnetic (EM) fields can cause malfunctions in many types of medical devices [1,2,3,4,5,6].  These problem reports are probably greatly underestimated due to the invisible nature of EM fields and the lack of the medical community’s familiarity of EMI.  Examples of devices that have malfunctioned, causing injuries and deaths include ambulatory and body-mounted medical devices (implanted cardiac and neural stimulators), home-use devices (apnea monitors), and clinically-used devices (infusion pumps, ventilators).  This paper will summarize, briefly, radiated EMI of medical devices and the factors that affect the severity of this interference. It is important to understand the parameters of EM fields that influence the degree of radiated interference.

THE MOST IMPORTANT CHARACTERISTICS OF RADIATED EM FIELDS THAT AFFECT EMI

Electromagnetic (EM) fields are made up of separate electric and magnetic field components.  These fields are produced by electrical circuits and devices carrying voltages and currents.  Sources of EM fields that cause EMI include low-frequency AC power wiring in the home or hospital, radiofrequency signals from distant transmitters, and potentially strong signals from nearby medical equipment, and wireless communication devices (e.g. handheld cellular phones or handheld transceivers).  These fields “radiate” (travel though the air or other media) and induce voltages and currents in electrical conductors at distant locations that are exposed to the fields.

Exposure field strengths

The strengths of the electric and magnetic fields in the region containing the medical device are critical factors in determining whether EMI will occur.  These field strengths are expressed in terms of Volts/meter for the electric field and Amps/meter (or Tesla) for the magnetic field.  Field strengths diminish as the separation from the source increases.  The exact relationship between distance and field strength is complex and differs for each of several types of general sources of EMI, depending on the radiating antenna, transmitter power, and frequency.  In general, at distances greater than a wavelength (see below), these fields fall off in inverse proportion to distance.  For example, a handheld wireless (cellular) phone operating at 835 MHz produced the electric field strengths shown in Table 1.  These were measured at various distances from the phone, in a horizontal plane, in an anechoic chamber in the author’s lab.  The measured results agree well with theory (equation 1).  For reference, the most prominent international medical device EMI standard [7] calls for critical medical devices to be immune to field strengths of 10 V/m for frequencies from 80 to 2500 MHz.

TABLE I   Measured field strength versus distance from a wireless phone

Distance (cm)

E (V/m)

     50

10.1

 100

5.0

 200  

2.6 

 

E = k/d

(1)

where:

E = Electric Field Strength (V/m)
d = distance (cm)

Frequency

The degree of EMI experienced by a medical device is related to the frequency (in Hertz – Hz) of the EM fields.  Certain medical devices exhibit EMI sensitivity to a narrow range of frequencies [4].  This is due to “resonance effects” in the electronics or in the cables connecting electrodes or other devices to circuitry of the medical device experiencing EMI.  In addition, the wavelength of the electromagnetic field affects the amount of EMI experienced by a medical device.  The wavelength in meters (L) is related to the frequency as indicated in equation 2 where f = frequency in Hz

 

L = 3x10 8 / f

(2)

Induced voltage

The voltage induced in a medical device’s electronic circuitry causes EMI and depends on several factors.  Consider the example of a monitoring device connected to a patient via an unshielded cable and exposed to a uniform electric field (E) (fig. 1).  A voltage is induced between the case of the device enclosing the electronics and the proximal end of the cable where it enters the device.  The voltage can be calculated using equation 3, and is proportional to the E field strength and the extended length of the cable

 

VE max = 0.5 E L

(3)

The maximum induced voltage (VE max) exists when the E-field vector is aligned with the length (L) of the wire (see fig.1).


Figure 1. Device and cable exposed to an electric field

A second example is presented for an implanted device with unipolar leads exposed to a uniform magnetic flux density (B), a voltage is induced between the case (can) enclosing the electronics, and the end of the lead inside the case. The voltage induced is proportional to B, the area (A) of a closed loop, and the frequency (f) as indicated in equation 4.

 

VB max  = f  B A

(4)

Figure 2. Implanted device and unipolar lead exposed to a magnetic field

The closed loop is formed by the patient leads plus the conductive tissue between the distal lead tip and the case of the implanted device (see fig. 2).  The maximum magnetically induced voltage (VB max) exists when the B field vector is perpendicular to the closed loop.  This relationship is valid at frequencies whose wavelength is large compared to the size of the medical device and its leads.               

The use of bipolar stimulation leads can reduce the induced voltages greatly for both electric and magnetic fields. 

Modulation

The amplitude modulation imposed on an EM field affects the field’s ability to cause EMI.  Many medical devices are designed to be particularly sensitive to electrical signals that lie in a certain “physiological passband”.  This passband is comprised of a set of frequencies corresponding to endogenous signals in the body (e.g. 0.1 Hz to 500 Hz for ECG, or 1 Hz to several kHz for EMG).  Electromagnetic fields with modulation frequencies that lie within the physiological passband may be able to interfere with the medical device, even if the fundamental frequency (or carrier of frequency) of the emitted field is much higher than the frequencies in the passband.  Modulation can also be specified in terms of the time domain.  For example, worst case EMI testing of cardiac pacemakers from cellular phones and other wireless communications devices EMI is specified using radiofrequency (RF) fields pulsed on for 25 ms at 500 ms intervals [8].

Direct stimulation or heating of a patient’s neural and muscular tissues

Currents and voltages can be “injected” directly into electrically excitable tissues by external magnetic fields if a wire or other lead is implanted in these tissues.  The external fields induce currents in wires such as electrodes from an implanted or external medical device.  Magnetic fields such as those emitted by “security systems” have caused a number of serious incidents of direct stimulation in neural tissues [2]. Strong 27 MHz magnetic fields from shortwave diathermy applied to the head have lead to deaths of patients with implanted deep brain stimulators due to extreme heating (to 87o C) induced by the fields.  High induced currents were concentrated at the tip of the implanted lead.

SOURCES OF RADIATED EMI

The U.S. Food and Drug Administration (FDA) has identified a number of sources of radiated electromagnetic fields, each of which has caused numerous cases of interference in medical devices and patient injuries.  Measured fields (pulsed or continuous wave) emitted by typical versions of these sources are shown in Table 2. The relative significance of the threat of EMI from each source is not listed.  This is because the total number of reports of EMI from any of these sources is generally under a few dozen in the FDA problem databases and other sources of information known to the author.   

Security systems

Electronic Article Surveillance Systems (EASS) for anti-theft use, and walk-through metal detectors emit EM fields that cause interference [9][10].  EASS and metal detectors usually consist of a transmit coil and a receive coil on opposite sides of a gate or portal.  Persons walk between the coils.  The transmit coil emits magnetic fields typically in the range of a few hundred Hz to a few MHz.

Wireless phones and handheld transmitters

This category includes cellular phones that are used by hundreds of millions of people worldwide.  Also included are higher power hand-held transmitters such as CB radios and portable transmitter/receivers used by security personnel. 

Magnetic Resonance Imaging (MRI)

MRI systems expose patients, operating staff, and nearby medical devices to intense pulsed and static magnetic fields.  This is especially true for “Open” MRI where a clinician performs interventional procedures very close to the system. Use of MRI is contraindicated for patients with metallic implants.

Diathermy

Shortwave diathermy devices radiate up to 50 Watts of power at 27 MHz to purposely heat selected tissues of a patient.  Use of diathermy is contraindicated for wearers of medical implants because of the known effect of extreme heating near the implant.

Electrosurgery units (ESU)

These medical devices use radiofrequency voltages to cut and cauterize tissues via electrodes contacting a patient’s tissues. ESUs cause malfunctions in patient monitoring and other instruments in the operating room.  While ESUs radiate strong electric and magnetic fields at 500 kHz and above, most EMI is due to conduction of their signals between instruments attached to the patient via wires. 

Radio and TV Transmitters

Stationary transmitter towers radiate many thousands of Watts of power to send their signal to distant receivers.  Field strengths close to the transmitter (less than 100 meters) can exceed levels that medical devices are intended to withstand.

MINIMIZING EMI SUSCEPTIBILITY THROUGH PROPER EMC DESIGN, STANDARDS, AND USER CONTROLS

Device design considerations

Medical devices can usually be made much less susceptible to EMI if various considerations are incorporated during the initial design of the device.  Shielding the electronics so that they are completely encased in an electrically-conductive enclosure can make medical devices highly immune to EMI.  Special filters are needed to keep EM voltages from being conducted through the shield from external input or output leads to the device.  For example, implanted cardioverter defibrillators and neural stimulators are shielded by their sealed metallic cases, and have special electronic filters at the input header to keep high frequency energy that may be present on the external leads from entering the interior circuitry.  However, it is difficult to make filters that are effective at low frequencies while not compromising a medical device's performance and safety requirements (maximum allowable leakage current from the lead to the case).  In addition to hardware, well-designed software in these devices can recognize and minimize the effects of EMI. All of these changes have been used to produce high immunity to interference from EMI sources such as cellular phones and certain other radio transmitters.

Contraindications, warnings, and proper use of medical devices near EM emitters

Another approach to reduce the risks from EMI in medical devices is to issue warnings, recommended practices, or contraindications.  For medical procedures that involve the use of strong EM emitters, such as MRI examinations, patients with implanted medical devices are not allowed to be examined with these emitters. Outside the clinical environment, the clinician and others can recommend procedures to patients who use medical devices that may be subject to EMI.  For example, the U.S. FDA identified simple precautions that wireless phone users who have implanted cardiac pacemakers may take [11].  These include establishing a minimum distance between the pacemaker and the phone (at least 6 inches or 15 cm), holding the phone to the ear opposite the side of the body where the pacemaker is implanted, and not carrying the phone in a shirt or jacket pocket directly over the pacemaker.

As a final example, the U.S. Food and Drug Administration worked with manufacturers of anti-theft security devices to address an issue involving existing users of implantable cardiac defibrillators (ICDs) and pacemakers [12].  They developed a letter recommending that all manufacturers of electronic anti-theft systems develop labeling or signage to post on or near all new and currently installed security systems.  The signs should indicate that an electronic anti-theft system is in use.  Such labeling or signage would also tell medical device implant wearers to avoid lingering around or leaning against systems that may affect their implanted electronic medical devices.  An FDA letter to physicians also recommended advising patients to not to lean on or linger in the gates when passing through.

DISCUSSION AND CONCLUSIONS 

Medical device malfunctions have been induced by nearby clinical sources of radiated EMI (such as ESUs) for many decades.  Users and manufacturers of medical devices accommodated these problems through equipment redesign and usage practices (e.g. separating sensitive equipment from the ESU leads).  Use of MRI and shortwave diathermy is contraindicated for wearers of most implanted metallic objects, including electronic devices, because of the known effect of interference and extreme heating near the implant.  However, a few patient injuries and deaths still occur in clinical settings because of disregard of the contraindications. Other, less obvious sources of EMI are now affecting devices occasionally in the clinical environment and in ambulatory settings.  In perspective, the number of EMI incidents is very small even considering the vast amount of under-reporting that exists.  However, new EMI problems need to be addressed as the world becomes more “wireless”.  By designing new medical devices for immunity to EMI, the risks to patients can be controlled.  Finally, existing medical devices can be protected from EMI by establishing awareness and good usage practices among clinicians who use and prescribe them.

REFERENCES

[1] J. Silberberg, “Electronic Medical Devices and EMI”, Compliance Engineering, 1996 Reference Guide, pp. D-14-D21, 1996

[2] D. Witters, "Examining Potential EMI Between Medical Devices and Electronic Security Systems", Medical Device and Diagnostic Industry, pp. 196 – 204, Jan. 2000

[3] W. Kainz, G. Neubauer, F. Aleschet, et al., “Electromagnetic Compatibility of Electronic Implants -- Review of the Literature”, Weiner Klin Wochenschr (the Middle European Journal of Medicine), Vol. 113 No. 23 pp. 903 – 914, 2001

[4] P. Ruggera,  E. O’Bryan, “Studies of Apnea Monitor Radiofrequency Electromagnetic Interference,” Proc. Ann. Internat. Conf. IEEE Engr. in Med. and Biol. Soc., Vol. 13, No. 4 , pp. 1641-1643, 1991

[5] M. McIvor, “Environmental Electromagnetic Interference from Electronic Article Surveillance devices: Interactions with an ICD”, PACE Vol.  18, pp. 2229-2230, 1995

[6] V. Barbaro et al., “Do European GSM Mobile Cellular Phones Pose a Potential Risk to Pacemaker Patients?”  PACE vol. 18, pp. 1218-1224, 1995

[7] IEC 60601-1-2 (2001).  General Requirements for Safety: Electromagnetic Compatibility -- Requirements and Tests (General).  International Electrotechnical Commission.

[8] AAMI - “Active Implantable Medical Devices -Electromagnetic Compatibility - EMC Test Protocols for Implantable Cardiac Pacemakers and Implantable Cardioverter Defibrillators”, AAMI/CDV-1 PC69, Association for the Advancement of Medical Instrumentation

[9] J. Casamento, “Characterizing Electromagnetic Fields of Common Electronic Article Surveillance Systems,” Compliance Engineering, pp. 42-52, September/October 1999,

[10] U.S. Food and Drug Administration (FDA), “Important Information on Anti-Theft and Metal Detector Systems and Pacemakers, ICDs, and Spinal Cord Stimulators”, FDA, Center for Devices and Radiological Health, Sep. 1998

[11] U.S. Food and Drug Administration, “Update on Cellular Phone Interference with Cardiac Pacemakers”, FDA, Center for Devices and Radiological Health, Sep. 1997

[12] U.S. Food and Drug Administration (FDA), “Labeling for Electronic Anti-Theft Systems”, FDA, Center for Devices and Radiological Health, Aug.  2000

[13] P. Ruggera, “Near-Field Measurements of RF Fields”, in Symposium on Biological Effects and Measurement of Radio-Frequency/Microwaves, Food and Drug Administration Conference Proceedings, Feb. 1977 pp. 104-116

Acknowledgement:  The author recognizes the significant contributions of Jon Casamento, Paul Ruggera, Donald Witters and other members of the Electrophysics Branch of FDA/CDRH in studying and measuring medical device EMI.

Table 2. Common Sources of  Medical Device EMI

 Emitter 1

Distance where

Frequency

E field
(V/m)

H field
(µT)

Modulation 2

Medical Devices Affected

Anti-theft devices

Center of portal

200 Hz – 60 kHz

Low 1

60 - 125

CW and pulsed

Impl. Cardiac and Neural Stimulator

Shortwave Diathermy

15 cm from applicator

27 MHz

> 300

> 1

Pulsed or CW

Deep Brain Stimulator, other implants

Electrosurgery Devices

16 cm from cable

450 – 550 kHz

> 500

> 0.5

Pulsed or CW

Operating room instruments

Hand-Held Transceivers  (Amateur and Guard Radio)

0.5 m

150 – 450 MHz

> 25

Low 1

CW

Monitoring equipment

Metal Detectors (walk-through)

Center of portal

200 Hz –
1 MHz

Low 1

5-28

Pulsed or CW

Implanted Neural Stimulator

MRI  (RF coil)

Center of coil

60 MHz or higher

Low 1

10  

Pulsed

Various implants

MRI (gradient coil)

Center of coil

0.5 – 2 KHz

Low 1

0.01T

Pulsed

Various implants

Transmitter towers (AM Radio)

100 m

0.5 – 1.7 MHz  (U.S.)

~ 20

 Low 1

CW

Patient monitors,  medical telemetry

Transmitter towers (TV)

100 m

50 – 800 MHz (U.S)

~ 20

Low 1

CW

Apnea monitor, patient monitors,  medical telemetry

Wireless (cellular) phone

0.5 m

824 - 849 MHz (U.S.)

> 10

Low 1

CW and Pulsed

Impl. Cardiac Pacer/defib.


    Notes:   (1) Not a significant source of EMI;    (2) CW = Continuous Wave (non pulsed)      

 

previous paper table of contents next paper

© International Society for Bioelectromagnetism