Effects of magnetic resonance imaging on cardiac pacemakers and electrodes Stephan Achenbach, MD, Werner Moshage, MD, Bj6rn Diem, MD, Tobias Bieberle, a Volker Schibgilla, MD, and Kurt Bac]hmann, MD Erlangen, Germany

In phantom studies we investigated the effects of magnetic resonance imaging (MRI) on pacemakers and electrodes. Twenty-five electrodes were exposed to MRI in a 1.5T scanner with continuous registration of the temperature at the electrode tip. Eleven pacemakers (five single chamber and six dual chamber) were exposed to MRI. Pacemaker output was monitored to detect malfunction in V O O / D O 0 and VVI/DDD modes. A temperature increase at the electrode tip of up to 63.1 ° C was observed during 90 seconds of scanning. In seven electrodes the temperature increase exceeded 15 ° C. Although no pacemaker malfunctions were observed in asynchronous pacing mode (VOO/DO0), inhibition and rapid pacing were observed during spin-echo imaging if the pacemakers were set to VVI or DDD mode. Pacemaker function was not impaired during scanning with gradient-echo sequences. Next to pacemaker dysfunction, electrode heating has to be considered a possible adverse effect when exposing patients with pacemakers to MRI. (Am Heart J 1997;134:467-73.)

Implanted cardiac pacemakers are commonly regarded to constitute a contraindication to magnetic resonance imaging (MRI). 1-5 All the same, there has b e e n a number of recent publications reporting successful MRI in patients w h o carried pacemakers. 6-~2 In spite of reports of a patient w h o died during MRI and rapid cardiac pacing observed in another patient, 8 some authors conclude that MRI can be peffomed safely in patients with implanted permanent cardiac pacemakers. 8,11-13 The effects of MRI on cardiac pacemakers are complex and difficult to investigate systematically. Several theoretic and p h a n t o m studies have looked at t h e behavior of pacemakers w h e n e x p o s e d to the strong magnetic and rapidly changing electromagnetic fields of MRI scanners. 13-18 Again, some of the authors concluded that MRI raay be performed safely in patients with pacemakers. However, the results were heterogeneous, partly because of the high number of possible combinations of pacemaker models, MRI scanners with various field strengths, and different imaging sequences that can be applied. The effects of MRI on implantable pacemaker electrodes has never b e e n studied systematically. Therefore w e attempted to investigate several commonly used pacemaker models From Medizinische Klinik II mit Poliklinik and the °lnstitut f~JrBiomedizinische Technik, University of Erlangen-NiJrnberg. Received for publication Oct. 4, 1996; acceptedApril 24, 199Z Reprint requests: Stephan Achenbach, MD, Medizinische Klinik II mit Poliklinik, Universit~t Erlangen-NOrnberg, C)stliche Stadtmauerstr. 29, D-91054 Erlangen,

Germany. Copyright © 1997 by Mosby Year-Book, Inc. 0002-8703/97/$5.00 + 0 4/1/82873

in various pacing modes, as well as different electrode types, as to their behavior w h e n e x p o s e d to imaging sequences used for routine investigations in a 1.5T MRI unit. During MRI scanning, rapidly changing magnetic and electromagnetic fields, next to a strong static magnetic field, are generated. Their interaction with metallic implants such as cardiac pacemakers could lead to several adverse effects: The static magnetic field could cause dislocation of the pacemaker and electrodes, should they contain ferromagnetic material. Because of the electromagnetic fields, pacemaker electronics could be destroyed or the programmed settings could be altered and, through induction of currents in the electrodes, sensing and triggering with consequent inhibition or rapid pacing, as well as heating of metallic components, could occur. By exposing 11 pacemakers and 25 electrodes in different settings to commonly used MRI sequences, w e studied these possible adverse effects.

Methods Magnetic resonance imaging The phantom studies were performed in a con~mercial Magnetom 1.5T (Siemens) magnetic resonance system installed at our hospital. To test pacemaker electrodes, untriggered Tl-weighted spin-echo sequences with an echo time (TE) of 20 msec and repetition time (TR) of 0.3 seconds, slice thickness of 8 mm, gap of 2 mm, a field of view of 400 mm, and a matrix 128 x 256 were chosen. Continuous scanning was performed for 90 seconds. To test pacemaker performance during MRI scanning, three

468

Achenbach et al.

M a x i m u m temperature increase (dT, o C after 90 sec) Electrode CPI 4185 CP14161 Biotronic TIR 60-UP Biotronic TIJ 53-UP Medtronic Capsure SP 4023 Medtronic Target Tip 4581 Medtronic Target Tip 4081 Biotronic SD 60-UP (V126) Biotronic EL 004128 Baxter 97-130-5F Biotronic Multicath 3 Biotronic TIJ 53-BP Biotronic TIR 60-BP Biotronic SD 60-BP Biotranic 385139 CPI 4268 CP14270 Medtranic Capsure SP 4524 Medtronic Capsure SO 4024 Medtronic Target Tip 4582 Medtronic AJ 049896V Siemens-Pacesetter 1450T Pacesetter 1188T Teletronics 033-301 Teletronics 033-856

Bipolar/ unipolar unl unl unl unl unl unl unl

unl unl bi (trap) bi (tmp) bi bi

bi bi bi bi bi bi bi bi bi bi bi bi

Length (cm)

Air, no PM

Air, connected to PM

NaCI solution connected to PM

59 59 59 51.5 58 53 65 59.5 60

16.9 20.4 16.6 0.9 9.3 -0.3 26.9 32.5 0

2.8 2.9 3.0 2.2 5.7 3.1 3.1 3.7 0

0.9 0.8 0.9 1.0 3.0 0.6 0.6 3.7 1.1

*

NA

NA

38.2 0.5 4.8 0.4 0.6 0.1 1.3 0.3 0.8 2.1 1.3 1.7 0,2 3.1 0,3

NA 0.1 0.0 -0.4 1.1 -0.1 2.3 0.8 2.0 0.4 3.3 1.2 -0.1 1.7 0.2

NA 0.5 0.7 0.3 4.9 0.4 4.4 8.9 1.2 3.8 2.4 1.1 0.1 0.9 1.4

116 114 51.5 58.5 59 59.5 52.5 52.5 53 58 53 58 58 52.5 58 48

Thetemperatureincreaseduring90 secondsof continuousscanningis givenfor threedifferentsellings:electrodeinsertedto the porcineheart,electrodeconnectedto a pacemaker and insertedto the heart,and dectrodeand pacemakersubmersedin NaCI solution. PM, Pacemaker;uni, unipdar;bi, bipolar;trap, temporarypacemakerelectrode;NA, not applicablebecausetemporarypacemakercould not be exposedto MRI. *Scanninginterrupledbecausetemperatureexceeded88.8° C (maximumrangeof thermometer).

different scanning sequences were chosen as used in routine applications and modified only to limit the overall scan time to 90 seconds: (1) untriggered Tl-weighted spin-echo sequence (TR 0.3 seconds, TE 20 msec, 8 m m slice thickness, 2 m m gap, FOV 400 mm, and matrix 128 x 256); (2) electrocardiographic-triggered Tlweighted spin-echo sequence (TR 0.7 seconds, TE 20 msec, 8 m m slice thickness, gap 2 mm, FOV 400 mm, and matrix 128 x 256); and (3) electrocardiographic-triggered gradient-echo sequence (TR 0.7 seconds, TE 10 msec, flip angle 40 degrees, 8 m m slice thickness, 2 m m gap, FOV 400 ram, and matrix 128 x 256).

Electrode testing To test the heating of pacemaker electrodes during MRI, 25 different electrodes were exposed to MRI with continuous registration of the temperature at the electrode tip. Two temporary electrodes and 23 implantable electrodes (nine unipolar and 14 bipolar) were investigated (Table I). To achieve conditions similar to those of the in vivo situation, the electrodes were connected to an isolated porcine heart

by inserting them into a deep cut in the left ventricular myocardium of the heart. The impedances of the electrodes were measured and were around 1000 m in all cases. An optical temperature sensor was connected to the electrode tip. Through an optical conductor (length 10 m), it was connected to the registration electronics outside the magnetic resonance scanning room (Luxtron 1000B Biomedical Fluoroptic Thermometer; Polytec). In a series of tests it was determined that maximum electrode heating occurred if the electrodes were arranged in a circular manner and laid fiat on the scanning table (electromagnetic field gradient perpendicular to electrode loop). Consequently, all electrodes were tested by exposing them to MRI scanning in the configuration of a circle with a diameter of 16 cm, which was oriented parallel to the plane of the MRI bed. Every electrode was tested without being connected to a pacemaker and while connected to a pacemaker (Paragon, SSI mode; Siemens-Elema), both in air and submersed in 0.9% NaC1 solution. Scanning sequences as described above were performed for 90 seconds with

American HeartJournal September 1997

Achenbach et ak

100 Pacemaker

Single/dual chamber

°C

Modes tested 80

Telectronics Meta 1206 Telectronics Simplex Biotronik Pikos CPI Vigor 460 CPI Vigor 1130 TelectronicsMeta 1254 Biotronik Ergos 03 Biotronik Physios01 CPI Delta T Pacesetter Paragon II Pacesetter Synchrony I!1

Single Single Single Single Single Dual Dual Dual Dual Dual Dual

VOO* VOO* VOO, VVI VOO, VVI VOO, VVI DOO* VOOI VVI, DOO, DDD VOO, WI, DOO, DDD VO©, VII, DOO, DDD VOO, VII, DOO, DDD VOO, VII, DOO, DDD

* Magneteffectcouldnotbe suppressedby programming.

continuous monitoring of the temperature at the electrode tip. They 'were interrupted if the temperature exceeded 88.8 ° C, the maximum range of the temperature measurement device used. After scanning, the porcine heart was examined for visible lesions.

Pacemaker testing Five single-chamber and six dual-chamber pacemakers were included in the investigation (Table II). If possible, the pacemakers were programmed to suppress the magnet function (mode switch to VOO pacing at a fixed rate when exposed to a static magnetic field). All pacemakers were programmed in VOO and VVI modes (single-chamber pacemakers) or VOO, VVI, DOO, and DDD modes (dualchamber pacemakers), respectively. Pacing rates were programmed to be betweeen 60 and 75 beats/min. The pacemakers were connected to an isolated porcine heart through commonly used atrial or ventricular electrodes (Medtromc Target Tip 4582 [Medtronic Inc., Minneapolis, Minn.] or Biotronik SD 60-BP), as described previously. The impedances were measured to be between 1000 and 1200 m. Bipolar sensing and pacing configurations were programmed for the atrial and ventricular electrodes. The atrial sensing threshold was programmed to 1.5 mV and the ventricular sensing threshold to 3.0 inV. The atrial and ventricular pacing amplitudes were set at 4 to 5 V. The upper frequency rate in DDD pacing was set to 150 beats/min. With the MRI system's electrocardiographic monitoring equipment, electric recordings from the surface of the isolated heart were registered continuously to monitor pacemaker output. Because the pacemakers had a stimulus duration of only 0.5 to 1.5 msec, their stimuli would have been undetectable in. these electrocardiographic registrations during MRI scanning, because low-pass filters are used to suppress high-frequency artifacts from the electrocardiographic tracings. Therefore a combination of a diode and a

AmericankieartJournaJ Volume 134,Number3

SJ

60

4O

2O 10

20

30

'

40

50

60

70

80

go

Recordings of temperature at electrode tip during 90 seconds of MRI for seven electrodes in which temperature increase of more than 15 ° C was observed. (All measurements were in air without pacemaker connected. )

capacitor (capacity 1 I-tF) was connected to the pacemakers, parallel to the electrode. Thus a wider pacemaker signal of 3 to 4 msec was obtained. This guaranteed reliable registration of pacemaker signals and electrocardiographic triggering of the MR1 scanner to the pacemaker output. The electrocardiographic traces recorded during MRI scanning in the different pacing modes were evaluated to detect sensing and pacing defects and changes in pacing rate. After scanning, the pacemakers' settings were compared with those programmed before the MRI scan and all pacemaker functions were checked thoroughly to detect possible changes in pacemaker settings or damage to the electronics.

Results As r e p o r t e d in prior publications, 17,18 forces acting o n the p a c e m a k e r s and electrodes c o u l d be felt o n e x p o s i n g t h e m to the MRI system's static magnetic field. In t w o p a c e m a k e r s (Me m 1206 and Meta 1254, Telectronics Pacing Systems), the ferromagnetic forces w e r e strong e n o u g h to cause the p a c e m a k e r to m o v e w h e n placed o n the MRI bed. In all other p a c e m a k e r s and in all electrodes, h o w e v e r , the forces w e r e too w e a k to result in dislocation of the e q u i p m e n t .

Electrodes The primary aim of the investigation was to detect heating of the electrodes during MRI b y m e a n s of an optical temperature p r o b e c o n n e c t e d to the electrode tip. Reproducibly, an increase in the temperature m e a s u r e d at the electrode tip that e x c e e d e d 15 ° C was r e c o r d e d in s e v e n of the 25 electrodes tested (Fig. 1).

469

470

Achenbach et al,

Electrode type Temporary Unipolar Bipolar

Setup No PM, air No PM, air PM, air PM, NaCI No PM, air PM, air PM, NaCI

Mean dT (o C)

Maximum dT (o C)

50.7-+ 1Z6 13.7+ 12.0 2.9_+2.0 1.5 _+ 1.1 1.3 _+ 1.3 0.9 _+ 1.2 2.2_+2.5

63.1 32.5 5.7 3.7 4.8 3.3 8.9

PM, Pacemaker.

The maximum temperature that could be documented was 88.8 ° C, the upper limit of the themlometer's measurement range. This corresponded to a temperature increase of 63.1 ° C, from a starting temperature of 25.7 ° C. The two electrodes with the highest increase in temperature were temporary pacing electrodes (Table I). Severe burns occurred at the insertion site of the electrodes; in one case confirmation was obtained in a histologic specimen. Only in four electrodes was the temperature increased less than 1.0 ° C for all of the three configurations tested, whereas the remainig 14 electrodes showed a temperature increase between 1.0 ° and 15 ° C (Table I). The average temperature increase of the electrodes at the electrode tip tended to be less pronounced when the electrodes were connected to a pacemaker and further attenuated when both the pacemaker and electrode were submersed in a tank filled with saline solution (Tables I and III). In general, heating of unipolar electrodes was stronger than that of bipolar electrodes. Pacemakers In all 11 pacemakers studied, MRI, with the scanning sequences described, showed no effect on the programmed pacemaker settings. Neither the setting nor pacemaker programmability was changed when checked after MRI. In all pacemakers the reed switch was operated by the static magnetic field. This resulted in asynchronous pacing at the preprogrammed rate when the pacemaker was attached to the MRI unit. None of the pacemakers displayed a pacing dysfunction when programmed to VOO or DOO mode (asynchronous mode) and exposed to MRI, regardless of the scanning sequence used. In three pacemakers the magnet effect could not be

switched off by programming before MR[ (Meta 1206, Meta 1254, and Simplex, Telectronics). For this reason, these three pacemakers could not be studied in modes other than VOO or DOO. In the remaining three singlechamber pacemakers and five dual-chamber pacemakers, severe effects on the pacing function were observed after programming the pacemakers to DDD or VVI mode. The effects were observed in all of the pacemaker models, but they varied according to the pacemaker settings and applied MRI sequences. Untriggered spin-echo sequences. In all singlechamber and dual-chamber pacemakers programmed to VVI mode, inhibition was observed on exposing the pacemakers to MRI with the magnet function inactivated. In one case the pacemaker was inhibited during the complete duration of the MRI acquisition. In the other models the duration of inhibition ranged from 1.4 to 9 seconds. Complex effects were observed when dual-chamber pacemakers set to DDD pacing were exposed to MRI scanning: Complete inhibition of the atrial and ventricular lead and isolated sensing/inhibition of one lead was observed (Fig. 2). Atrial triggering with consequent rapid ventricular pacing at the upper frequency limit (150 beats/rain) was also documented (Fig. 3).

Electrocardiographically triggered spin-echo sequences. The same effects as in untriggered imaging sequences were observed during electrocardiographically triggered spin-echo sequences. However, because these scanning sequences are interrupted after 700 msec (the preset TR value) to wait for the next electrocardiographic trigger, total inhibition was not observed. The effect resulted in a decreased pacing frequency of as low as 27 beats/min (Fig. 4). Again, atrial triggering was observed in DDD pacemakers, resulting in ventricular pacing at the upper programmed frequency limit.

American Heart Journal September 1997

Achenbach et al.

Dual-chamber pacemaker, DDD pacing mode. After onset of spin-echo sequence (arrow), complete inhibition occurs.After 2.4 seconds,single ventricular spike follows, caused by atrial sensing/triggering.

Rapid ventricular pacing at upper frequency limit of 150/min after onset of MRI (arrow, untriggered spin-echosequence), with pacemaker in DDD mode. After scanning is interrupted (arrowhead), normal atrial and ventricular pacing continuesat 75/min.

Eleetrocardiographically triggered gradient-echo sequences. Pacemaker functions remained completely uninfluenced by MRI when electrocardiographically triggered gradient-echo sequences were used.

Discussion During diagnostic MRI, several forms of scanning sequences (spin-echo sequences or gradient-echo sequences) are applied. All are based on the electromagnetic deflection of hydrogen atoms that were aligned in a strong static magnetic field and subsequent measurement of the energy emitted. They differ only in the magrfitude, spatial orientation, and temporal coordination of the applied fields. The static magnetic field, the rapidly changing magnetic field (gradient field, 100 to 200 Hz), and the electromagnetic radiofrequency field (63 MHz) interact with implanted pacemakers and electrodes. For this reason, patients with pacemakers have by general policy been excluded from MR[.1-4 All the same, several authors have reported MRI investigations of patients with implanted pacemakers. In a thorough search of the literature, we found reports of a total of 36 patients. Side effects, including one death, were reported in four of these patients (Table IV). Because the interactions between MRI scanning and pacemakers are complex and not yet fully understood, and the effects of MRI on electrodes have not been investigated, we investigated the effects of the static magnetic field and the electromagnetic fields on both pacemakers and electrodes in phantom studies.

Static magnetic field The static magnetic field can exert forces of up to 5N (about tenfold the pacemaker's weigh0 on pacemakers containing ferromagnetic materials. 18 Only two of the pacemakers in our study were moved when positioned freely on the MRI examination bed. The danger of

American Heartjournal Volume 134, Number 3

Slow ventricular pacing (37/min) during MRI with electrocardiographically triggered spin-echosequences. Normal rate (60/min) resumesafter termination of scanning

(arrow).

dislocation can be ruled out for the other models when implanted in a patient. Also, the static magnetic field usually operates the pacemaker's reed switch, causing asynchronous pacing, which, according to our results, seems to be the safest pacing mode during MRI. However, it is theoretically possible that the reed switch is not activated if the MRI static magnetic field is oriented perpendicular to the reed-switch axis. 18 In this case, the effects of inhibition or triggering that we could observe during pacing in VVI and DDD pacing modes may constitute a danger to the patient. Also, the operation of the reed switch itself and consequent asynchronous pacing have been reported to be the cause of ventricular fibrillation in a patient. 19 For this reason, no patient with a pacemaker should approach an MRI Unit without availability of a defibrillator.

Gradient magnetic field and radiofrequency field The effect of these rapidly changing electromagnetic fields on electrodes has been investigated systematically for the first time in our study. In our phantom experiments, exposing isolated pacemaker leads to MRI caused significant heating of the electrodes. The heating is caused by the induction of currents in the electrodes,

471

472

Achenbach et al.

Author Gimbel et al. 8

Johnson 11 Achenbach et al. 6 Alagena et al. 7 Iberer et al. 9 Inbar et al. 1° Lauck et al. 12

No. of patients studied

MR scanner field strength (T)

20

4 1 1 1 1 8

which act as antennas for the electromagnetic fields. We recorded maximum temperatures of almost 90° C, and myocardial necrosis could be demonstrated in histologic studies, but there were large differences between the different electrodes: only seven electrodes showed an increase in temperature of more than 15° C during 90 seconds of imaging. The effects were attenuated when a pacemaker was connected to the electrodes and the electrodes were submersed in saline solution. The in vivo effect might be even less pronounced because the electrodes are not arranged in a full loop, and the moving blood can be expected to reduce heating of the electrodes further by means of convection. Nevertheless, we believe that electrode heating constitutes an eminent danger for patients subjected to MRI: Because imaging usually extends for a longer time than our 90-second sequences (usually 4 to 10 minutes of uninterrupted scanning) and the electrode temperatures were rising continuously to the interruption of scanning (Fig. 1), the in vivo heating effects might be more pronounced than in our study. In addition, electrode heating is not detectable by monitoring and the effects such as an increase in pacing threshold, right ventricular or atrial perforation, or even reentry arrhythmias caused by tissue scars can occur long after imaging. Because heating varies considerably between different electrodes, it is mandatory that every electrode be tested in phantom studies before exposing any patient with an implanted electrode of that type to MRI. It must also be kept in mind that many patients carry isolated pacemaker leads left in place after pacemaker removal. In animal studies conducted to test the behavior of transesophageal pacing leads during MR[, heating with consequential necrosis has also been observed. 20 The postulated effects of the gradient field and radiofrequency fields on the pacemakers themselves

0.5-1.5

Not given 0.5 1.5 1.5 1.5 0.5

Adverse effects Discomfort at PM pocket (1 patient), rapid heart rate (1 patient), death in unmonitored patient (1 patient) Transient inhibition (1 patient)

include interference with pacemaker electronics , inhibition, and rapid pacing. 2-4 In all 11 pacemakers tested in our series, pacemaker electronics seemed to be unharmed by MRI: all pacemakers had unchanged programmability when checked after MR[, and the settings were not altered during the scanning procedure. Inhibition, caused by the induction of currents above sensing threshold in the pacemaker leads, was observed in all pacemakers set to VVI or DDD mode if spin-echo sequences were used. The same has been reported in previous studies of other pacemaker models. 14,15,17 Gradient-echo sequences, which require less highfrequency energy, showed no effect in our investigation. To avoid inhibition, setting the pacemaker to VOO, DOO, or, preferably, OOO mode seems necessary when attempting MR[. Rapid pacing can be the result of two mechanisms: In our study we observed rapid pacing at the upper frequency limit for short periods in all DDD pacemakers investigated. In our opinion, this effect is caused by induction of currents above sensing threshold in the atrial lead and consequent triggering of ventricular stimulation. Direct interference with the pacemaker electronics seems unlikely because the rapid pacing rate was always equal to the programmed frequency limit. However, in animal studies that Were conducted with different implantable pacemakers, pacing of up to 300 beats/min has been observed synchronized to the radiofrequency pulses and was attributed to interference with pacemaker electronics. 1416 , In summary, both our results and those published by other authors indicate that only under very special circumstances can MR[ be performed safely in patients with implanted cardiac pacemakers. Inhibition, rapid pacing, and induction of arrhythmias have been reported previously as possible adverse effects. They

American Heart Journal September 1997

Achenbach et al.

can be detected by adequate patient monitoring during MRI and can most probably be prevented by suitable scanning strategies. Heating of the electrodes, however, is a potentially harmful effect that has never before been investigated systematically and is especially dangerous because it cannot be detected by patient monitoring. In our studies only one magnetic resonance scanner was used, the field strength of which is typical for those used in diagnostic imaging (1.5T). It can be expected that adverse effects are less pronounced in scanners with a lower magnetic field strength, 16 but differences would be only quantitative, not qualitative. We agree with several previous publications that MRI should never be conducted in a patient with a pacemaker without the highest urgency, the pacemaker and electrodes identical to those implanted in the patient have to be tested in phantom studies, and results and observations with one pacemaker must never be extrapolated to other, even very similar, products.. It is our concern that reports of successful MRI in a limited group of patients with pacemakers may dilute the general policy of never exposing a patient with a pacemaker to MRI. Carelessness or reduced awareness of the potential dangers could cost a patient's life.

References 1. Abart J, Ganssen A. Safety aspects of MR imaging. Aktuelle Radiol 1995;5:376-84. 2. Heiken JP, Brown JJ. Manual of clinical MRI: a practical guide to conducting magnetic resonance imaging examination of the head and body. New York: Raven Press; 1991. 3. Moshley I. Safety and magnetic resonance imaging. Br Med j 1994;308:1181-2. 4. She%ck F, Kanal E. PoLitics,guidelines, and recommendations for MR imaging safety and patient management. J Magn Reson Imaging 1991 ;1:97-104.

American k-leartJournal Volume 134, Number 3

5. Zimmermann B, Faul D. Artifacts and hazards in NMR imaging due to metal implants and cardiac pacemakers. Diagn Imaging Chn Med 1984;53:53-6. 6. Achenbach S, Moshage W, Kuhn [, SchibgilJa V, Bachmann K. FallvorsteNung: Kernspintomographie bei einem Patienten mit Zweikammer4chrittmachersystem. Z Kardio[ 1995;84(suppl):119. Z Alagona P,Toote jC, Maniscalo B. Nuclear magnetic resonance imaging in a patient with a DDD pacemaker [letter]. Pace 1989;12:619. 8. Gimbel JR, Lorig RJ, Wikloff BL. Safe magnetic resonance imaging of pacemaker patients [abstract]. J Am Coil Cardio11995;25:]1A. 9. Lberer F,Justich E, Stenzl W. Nuclear magnetic resonance imaging of a patienl with implanted transvenous pacemaker. Herz 1987;7:196-9. 10. Inbar S, LarsonJ, Butt T. Case report: nuclear magnetic resonance imang in a patient with a DDD pacemaker. Am j Med Sci 1993;305:174-5. 11. johnson D. Magnetic resonance imaging effects and considerations with permanent cardiac pacemakers [abstract]. Pace 1994; 17(suppl):772. 12. Lauck G, Sommer T, Woike S, Luderitz B, Manz M. Magnetic resonance imaging in patients with implanted permanent pacemakers [abstract]. Pace 1995;18(suppl):1168. 13. Lauck G, von Smekal A, Wolke S, et al. Effects of nuclear magnetic resonance imaging on cardiac pacemakers. Pace 1995; 18:1549-55. 14. Erlebacher JA, Cahill PT, Pannizzo F, Knowles RJr. Effect of magnetic resonance imaging on DDD pacemakers. Am J Cardiol ]986;57: 437-40. 15. FetterJ, Aram G, Holmes DR, GrayJ. The effects of nuclear magnetic resonance imagers on external and implantable pulse generators. Pace 1984;7:720-Z 16. Hayes DL, Holmes DR, GrayJE. Effect of 1.5 tesla nuclear magnetic resonance scanner on implanted permanent pacemakers. J Am Coil Cardio11987; 10:782-6. 1Z Pavlicek W, Geisinger M, Castle L, et ak The effects of nuclear magnetic resonance imaging on patients with cardiac pacemakers. Radiology 1983; 147:149-53. 18. Tobisch RJ, Irnich W, Bachmann J, Batz L. Elektramagnetische Auswirkungen des Kernspintomographen auf den Herzschrittmacherpatienten. Biomed Tech (Ber]) 1993;38(suppl):435-Z 19. Seipel L, Bub E, Driwas S. KammerflJmmernbei FunktionsprLifungeines Demand-Schritlmachers. Deutsch Med Wochenschr 1975;100: 243944. 20. de Cock CC, Hofman MBM, van RossumAC, Visser FC, Visser CA. Safety and feasabili V of transoesophageal pacing during magnetic resonance imaging [abstract]. Eur HeartJ 1995;16(suppl):401.

473