EMF interference detection utilizing the recording feature of cardiac pacemakers
Tommi Alanko1 , Maria Tiikkaja2, Harri Lindholm3 , and Maila Hietanen4
1Finnish Institute of Occupational Health, Topeliuksenkatu 41 a A, 00250 Helsinki, Finland, email@example.com
2Finnish Institute of Occupational Health, firstname.lastname@example.org
3Finnish Institute of Occupational Health, email@example.com
4Finnish Institute of Occupational Health, firstname.lastname@example.org
Electromagnetic (EMF) interference with cardiac pacemakers may occur in various work environments. In the
case of interfering external signals, the pacemaker may misinterpret the signal as a heart-related problem and initiate
treatment procedures unnecessarily. We evaluated the applicability of the interference recording feature of cardiac
pacemakers to identify the interfering sources. The pacemakers were exposed to a wide variety of magnetic fields in a
sophisticated exposure setup in which the exposure was controlled by a computer programme. In cases where a
pacemaker experienced interference, the time of interference was compared with the magnetic field exposure schedule.
The study shows that interference recordings can be linked together with the exposure correctly, and it is possible to
differentiate the EMF-induced pacemaker interference from other types of interference. At workplaces, an EMF
recorder can be given to a pacemaker patient to wear for a few days, asking to keep a diary of his or her activities. After
an appropriate time, the EMF recordings are read and compared with the interrogated pacemaker data. The interference
source can be identified by combining the time of interference with the patient's diary activities.
The number of workers with cardiac pacemakers and implantable cardioverter defibrillators (ICD) is
increasing rapidly as the mean age of workers is rising. The European EMF Directive (2004/40/EC) requires employers
to consider specifically the factors that affect the health and safety of workers at particular risk, like those wearing a
pacemaker. This creates great challenges especially in industrial work environments, where high magnetic fields can be
found. Modern pacemakers record the heart's electric activity and the pacemaker's function during clinical disorders or
environmental interference as stored electrograms (EGM) [1, 2]. In the case of interfering external signals, the
pacemaker may misinterpret the signal as a heart-related problem, and initiate treatment procedures. Several cases of
false recognitions and treatments due to electromagnetic interferences have been reported [3-5]. The pacemaker records
the incidences with a timestamp, and these recordings have been used to connect the interferences with external
electromagnetic sources . In many cases, the connection can be made with situations and points in time when the
interference has occurred. In some studies, a real-time pacemaker monitoring, or ambulatory electrocardiogram (ECG)
recordings have been used to study how the pacemaker reacts to outside interferences [7-9]. Continuous monitoring by
telemetry is not feasible in many cases due to susceptibility to EMF interference [10, 11].
Separate measurements of the EMF strengths are time consuming and not applicable to ambulatory ECG type
measurements, in which the results are analysed after a longer time period to find the possible interference sources. We
investigated the possibility of connecting time-stamped EMF strength measurements (e.g. EMF logger) to the
pacemaker's recordings. To examine this, a Helmholtz-coil setup was built to produce the magnetic fields, which were
measured with an oscilloscope. This in vitro setup can easily be applied to a real exposure situation by replacing the
oscilloscope with a device that can record the measured EMF exposure with a time-stamp. A similar measurement
approach has previously been used to study the possible dysbalance of autonomic nervous system regulation in patients
with perceived electrical hypersensitivity, by using external ambulatory ECG recordings and magnetic field recorders
2. Materials and Methods
Low frequency magnetic fields for interference testing were produced by a setup consisting of Helmholtz coils
with 17 turns (diameter 74 cm, gap 37 cm), a waveform generator (33220A Function Generator / Arbitrary Waveform
978-1-4244-6051-9/11/$26.00 ©2011 IEEE
Generator, Agilent Technologies), a power amplifier (KEPCO 20-20 bipolar power amplifier, KEPCO) and a current
meter and current shunt ( Agilent 34411A Digital Multimeter and 34330A 30 A Current Shunt, Agilent Technologies,)
The tests were controlled by a computer programme developed in the Agilent VEE graphical programming environment
(Agilent VEE 7.5). The programme controlled the test sequences, waveforms and amplitudes. Magnetic fields were
continuously monitored and recorded using a virtual oscilloscope (PicoScope 3424, Pico Technology) during the tests.
The tested pacemaker was placed in a phantom container filled with 0.9% saline solution and centred in the
Helmholtz coils. The container dimensions were 225*290*55 mm. The pacemaker and the lead formed an area of 190
squared-cm, which corresponds to the worst-case anatomical configuration from an electromagnetic interference. The
phantom was placed inside the Helmholtz coils and measured in three orthogonal directions to take into account the
directional effects. The control programme directed the waveform generator during the test schedule.
It is possible to choose the waveform (e.g. sine, square, pulse, ramp) and the characteristics of certain
waveforms (e.g. rise time), the frequency and frequency sweeps (1 Hz - 200 kHz), the amplitude, the duration of the
exposure, and rest time between the exposures. The magnetic field strength can be reliably calculated from the
analytical formula for Helmholtz-coils. The test sequence length is not restricted by the exposure setup programme;
only by the pacemaker recording feature. If the number of recordings in the pacemaker is restricted, the maximum
number of test signals/interferences in one test run is also restricted. Before the test, the pacemaker was programmed to
the appropriate settings. The most sensitive settings are usually used to simulate the worst-case situation. When the test
begins, both the pacemaker and the oscilloscope record the events, and the corresponding time-stamped oscilloscope
recording is used to verify the actual magnetic field strength at the moment of interference.
The test protocols included several successive exposure situations. Exposures differed from each other by
magnetic field strength, frequency, waveform type and durations. In some cases the pacemakers experienced
interference; in others it did not. By comparing the protocol schedule and interrogated pacemaker information, it was
possible to reveal the conditions which induced the interference.
An example of EMF interference events is shown in Figure 1. The interference situation was as follows: during
a 10-sec magnetic field exposure (50 Hz, 520 ?T), an implantable cardioverter-defibrillator (ICD) experienced
interference lasting also 10 seconds. At the beginning of the exposure, the ICD programmed for ventricular rate
modulated pacing (VVIR) mode shifted into a ventricular demand (VVI) pacing noise mode. The pacemaker
categorised interference as ventricular tachycardia (VT), and anti-tachycardial pacing (ATP) was activated. Seven
seconds after the end of exposure, the ICD returned to normal mode and pacing. The ICD time tagging accuracy is
limited to a one minute precision, which is adequate to make a temporal distinction from different interference sources
in practical circumstances.
Figure 1. Simultaneous recordings of magnetic field exposure (above) and ICD's interference (below). The
applied magnetic field and the detected interference occur exactly during the same duration.
In Figure 2, another interference case is presented. An arrhythmia pacemaker was exposed to 60 Hz sinusoidal
magnetic field (B = 430 ?T). The pacemaker interpreted the interference as ventricular tachycardia and recorded the
Figure 2. Magnetic field interference interpreted by a pacemaker as tachycardia.
The findings of our study indicate that pacemaker EMF interference sources can be identified by connecting
pacemaker recordings to EMF recordings. When the measurement times of the two devices are correlated, it is possible
to recognise the EMF induced pacemaker interference correctly. The measurement result gives the best representation
of an effective electromagnetic field to the pacemaker when the recording device is close to the pacemaker. The
pacemaker EGM recording features, e.g. time-tagging accuracy, may limit the usability of this method. Nevertheless, as
the goal is to differentiate EMF-induced interferences from other disturbances, the existence/absence of EMF also gives
valuable information, and the recorder can be carried, for example, on a belt.
This assessment method can only be used in circumstances where it is known that the interference from the
EMF sources is so small that it does not pose a threat to the pacemaker wearer. The method can also be used when
moderate interference is perceived in medical checks, to exclude EMF interference from those caused by clinical
problems. An EMF recorder can be given to a pacemaker patient to wear for a few days, asking to keep a diary of his or
her activities. After an appropriate time, the EMF recordings are read and compared with the interrogated pacemaker
data. The location of the interference source can be revealed by combining the time of interference with the patient's
This work was supported by the Finnish Work Environment Fund [Grant number 107236].
1. K. Mortensen, V. Rudolph, S. Willems, and R.Ventura, "New developments in antibradycardic devices," Expert Rev
Med Devices, 4, 2007, pp.321-333.
2. B. H. Sarter, D. J. Callans, G. D. Gottlieb, D. S. Schwartzman, and F. E. Marchlinski, "Implantable defibrillator
diagnostic storage capabilities: evolution, current status, and future utilization", Pacing Clin. Electrophysiol, 21,1998,
3. A. Trigano, P. Deloy, O. Blandeau, and S. Levy, "Arc welding interference recorded by an implanted cardiac
pacemaker", Int J Cardiol,109, 2006, pp. 132-134.
4. P.A. Santucci, J. Haw, R.G. Trohman, and S. L. Pinski, "Interference with an implantable defibrillator by an
electronic antitheft-surveillance device", N Engl J Med, 339, 1998, pp. 1371-1374.
5. J.R. Gimbel, and J. W. Jr. Cox, "Electronic article surveillance systems and interactions with implantable cardiac
devices: risk of adverse interactions in public and commercial spaces", Mayo Clin Proc, 82, 2007, pp. 318-322.
6. T. Nagatomo, H. Abe, R. Kohno, T. Toyoshima, H. Fujimoto, S. Kondo, N. Kabashima, M. Takeuchi, M. Tamura,
M. Okazaki, and Y. Otsuji, "Electromagnetic interference with a bipolar pacemaker by an induction heating (IH) rice
cooker", Int Heart J, 50, 2009, pp. 133-137.
7. M. E. McIvor, J. Reddinger, E. Floden, and R. C. Sheppard, "Study of pacemaker and implantable cardioverter
defibrillator triggering by electronic article surveillance devices (SPICED TEAS)", Pacing Clin. Electrophysiol, 21,
1998, pp. 1847-186.
8. U.K. Wiegand, H. Bonnemeier, F. Bode, F. Eberhardt, J. K. Chun, H. A. Katus, and W. Peters, "Continuous Holter
telemetry of atrial electrograms and marker annotations using a common Holter recording system: impact on Holter
electrocardiogram interpretation in patients with dual chamber pacemakers", Pacing Clin Electrophysiol, 25, 2002, pp.
9. L. Toivonen, J. Valjus, M. Hongisto, and R. Metso, "The influence of elevated 50 Hz electric and magnetic fields on
implanted cardiac pacemakers: the role of the lead configuration and programming of the sensitivity", Pacing Clin
Electrophysiol,14, 1991, pp. 2114-2122.
10. G. Webster, L. Jordao, M. Martuscello, T. Mahajan, M. E. Alexander, F. Cecchin, J. K. Triedman, E. P. Walsh, and
C. I. Berul, "Digital music players cause interference with interrogation telemetry for pacemakers and implantable
cardioverter-defibrillators without affecting device function", Heart Rhythm,5, 2008, pp. 545-550.
11. K. Dyrda, and P. Khairy, "Interference between cellular telephones and implantable rhythm devices: a review on
recent papers", Indian Pacing Electrophysiol J,6, 2006, pp. 226-233.
12. M. Sandström, E. Lyskov, R. Hörnsten, K. Hansson, U. Wiklund, P. Rask, V. Klucharev, B. Stenberg, and P. Bjerle,
"Holter ECG monitoring in patients with perceived electrical hypersensitivity", Int J Psychophysiol,49, 2003, pp. 227-