Multifunctional Interventional Devices for MRI:A Combined Electrophysiology/MRI Catheter

Robert C. Susil, Christopher J. Yeung, Henry R. Halperin, Albert C. Lardo, andErgin Atalar*

The design and application of a two-wire electrophysiology (EP)catheter that simultaneously records the intracardiac electro-gram and receives the MR signal for active catheter tracking isdescribed. The catheter acts as a long loop receiver, allowingfor visualization of the entire catheter length while simulta-neously behaving as a traditional two-wire EP catheter, allow-ing for intracardiac electrogram recording and ablation. Theapplication of the device is demonstrated by simultaneouslytracking the catheter and recording the intracardiac electro-gram in canine models using 7 and 10 frame/sec real-timeimaging sequences. Using solely MR imaging, the entire cath-eter was visualized and guided from the jugular vein into thecardiac chambers, where the intracardiac electrogram was re-corded. By combining several functions in a single, simplestructure, the excellent tissue contrast and functional imagingcapabilities of MR can be used to improve the efficacy of EPinterventions. This catheter will facilitate MR-guided inter-ventions and demonstrates the design of multifunctional in-terventional devices for use in MRI. Magn Reson Med 47:594 – 600, 2002. © 2002 Wiley-Liss, Inc.

Key words: interventional MRI; active device tracking; cardiacelectrophysiology; catheter ablation

Electrophysiologic (EP) interventions are commonly per-formed to treat a variety of cardiac arrhythmias (1). Inthese procedures, RF ablation is used to eliminate focalatrial tachycardias, AV-nodal reentrant tachycardias, ac-cessory pathways, atrial flutter, and idiopathic ventriculartachycardia (2–5). Catheters are introduced through thefemoral vein and guided into the right heart using X-rayfluoroscopy (6). These catheters contain multiple wireleads, each terminating in its own electrode on the cathetersurface, which are used to record the intracardiac electro-gram (IEGM) (6) and also to deliver RF (kHz) energy forablation of arrhythmogenic cardiac tissue. By monitoringthe IEGM, arrhythmogenic cardiac tissue can be locatedand then ablated with RF (kHz) energy.

Currently, EP procedures are performed under X-rayfluoroscopy. While this modality has a high frame rate,small pixel size, and can easily localize catheters, it deliv-ers poor tissue contrast. Cardiologists can visualize onlythe catheters and faint silhouettes of the heart structure.Therefore, the IEGM is the main tool used to guide thecatheters to the proper cardiac tissue. In addition, X-ray

fluoroscopy only provides projection images and exposesthe patient and physician to ionizing radiation.

To overcome these limitations, MRI has been proposedas an alternative imaging modality for guiding and moni-toring EP procedures. MR offers several attractive featuresfor the EP interventionalist. First, the heart and endocar-dial landmarks can be visualized throughout the proce-dure. This anatomical information is crucial to identifyablation sites for arrhythmias such as atrial flutter andfibrillation. Second, cardiac motion and flow dynamicscan be monitored during the procedure. This enables acuteassessment of cardiac function as the intervention is per-formed. Most important, however, MRI allows for visual-ization of ablated tissue during the procedure. (7). Becauselesions are invisible under X-ray fluoroscopy, it is cur-rently difficult to discern whether tissue has been com-pletely ablated, is only temporarily stunned (i.e., subject toreversible thermal injury), or has not been treated at all (6).By actually visualizing the tissue lesion, MRI provides forpositive confirmation of tissue treatment. This is espe-cially important for procedures that require continuouslines of ablation, such as treatment for atrial flutter, atrialfibrillation, and isolation of pathologic tissue in the pul-monary veins. Imaging of ablated tissue may allow whatare now long and difficult procedures to be performed ona more routine and efficient basis.

However, to perform EP interventions under MRI, cath-eters that can perform multiple functions are needed. First,the catheters must be easily visualized and tracked in theMR images. Second, the catheters must record the IEGM(this is crucial for tissue localization and monitoring).Finally, it must be possible to deliver RF energy for abla-tion of the desired tissue near the catheter tip. Previouswork by our group has demonstrated the feasibility ofdelivering RF energy for cardiac ablation within the MRIscanner (7). Therefore, the present work will focus on onlythe first two functions.

In this study, we demonstrate the feasibility of using atwo-wire EP catheter to simultaneously record the IEGMand to receive the MR signal for active catheter tracking.The combined EP/MR antenna catheter serves simulta-neously as an MR antenna and as an EP catheter. At64 MHz, the catheter behaves as a long loop receiver,providing a local region of high signal that is ideal forcatheter tracking. Simultaneously, at lower frequencies thecatheter consists of two electrically isolated wire leads.These leads can be used to record EP signals and ablatetissue. By combining these functions in a single, simplestructure, the excellent tissue contrast and functional im-aging capabilities of MR can be used to improve the effi-cacy of EP interventions.

Departments of Biomedical Engineering, Radiology, and Medicine, JohnsHopkins University School of Medicine, Baltimore, Maryland.Grant sponsor: NIH; Grant numbers: RO1 HL57483; RO1 HL61672; RO1HL64795; RO1HL45683.*Correspondence to: Ergin Atalar, Ph.D., Johns Hopkins University, JohnsHopkins Outpatient Center Room 4241, 601 N. Caroline St., Baltimore, MD21287-0845. E-mail: eatalar@mri.jhu.eduReceived 2 July 2001; revised 23 October 2001; accepted 23 October 2001.

Magnetic Resonance in Medicine 47:594–600 (2002)DOI 10.1002/mrm.10088

© 2002 Wiley-Liss, Inc. 594

THEORY

The physical structure of a typical bipolar EP catheter isshown in Fig. 1a. A plastic shaft forms the body of the100-cm catheter, with two surface ring electrodes at thedistal end and two wire leads that run the length of thecatheter. The electrical structure is shown in Fig. 1b. Notethat each electrical lead terminates in its own surfaceelectrode. To track this type of catheter, it is advantageousto create a small, confined region of high SNR. The aim ofthis work, therefore, was to determine how to best accom-plish this with minimal modification of the existing cath-eter structure.

Numerous previous studies have shown that long, flex-ible loop antennas produce local regions of high signal andare therefore ideal for active catheter tracking (8–10). For acatheter structure with two electrical leads, this loop con-figuration can be produced by simply short-circuiting thedistal end of the catheter. While this solution may betheoretically ideal, it is an unacceptable solution for thecurrent problem. To use the catheter for EP interventionsand ablation, multiple, electrically isolated wire leads arenecessary. A short circuit at the catheter tip compromisesthis configuration.

To solve this problem, a capacitor can be placed at thedistal end of the catheter (Fig. 1c). In addition, RF chokes(parallel LC circuits tuned to 64 MHz) can be applied inseries with the surface electrodes to improve the safety ofthe EP catheter in MRI. To understand the function of thecatheter and the effects of these modifications, consider

the electrical topology of the device at both low- andhigh-frequency ranges. First, at low frequencies the capac-itor has a large impedance and behaves approximately asan open circuit. The RF chokes have low impedances andappear as short circuits. Therefore, the low-frequencystructure of the catheter is identical to that of the nativestructure (Fig. 1d). However, at high frequencies the struc-ture is quite different. The impedance of the capacitor athigh frequencies is small; it is modeled as a short circuit.The RF chokes have a very large impedance, behavingapproximately as open circuits. Rather than a two-leadcatheter, the catheter is now a long loop receiver (Fig. 1e),which is ideal for tracking. In addition, the two catheterelectrodes have been decoupled from the rest of the cir-cuit. Note that these two structures exist simultaneously,but at different frequency ranges. Therefore, this structureenables us to use the same leads for simultaneous MRimaging and IEGM recording.

MATERIALS AND METHODS

Catheter Implementation

To implement the combined function catheter design, acatheter was constructed using components from a gold-tipped, 7 french, MR-compatible two-electrode ablationcatheter (Bard, Murray Hill, NJ). Only the electrode tip andconnector of the commercial catheter were used. The twowire leads were replaced with 30 gauge insulated copperwire (0.25 mm wall thickness; AlphaWire Co., Elizabeth,NJ). By doing this, conductor separation was held constantat 1.3 mm, promoting uniform signal sensitivity along thecatheter shaft. The shaft of the commercial catheter wasreplaced with polyolefin tubing (3.36 mm diameter with0.5 mm wall thickness; AlphaWire). The final catheter sizewas 10 french (Fig. 2). A 500 pF chip capacitor (1.4 � 1.4 �1.45 mm) was placed between the catheter leads 2 cm fromthe catheter tip (American Technical Ceramics Corp., Hun-tington Station, NY). RF tip chokes were not included inthis implementation. As our objective is only to demon-strate the basic design and application of a combinedfunction catheter, the tip chokes are unnecessary at thispoint.

The catheter was connected to a tuning, matching, andsignal-splitting circuit (Fig. 2). Tuning and matching forthe MR signal was performed as described previously(8,11). The IEGM signal was passed through a set of RFfilters to prevent the electrogram receiver from loading theMR tuning/matching circuit (Fig. 2b). This signal was low-pass filtered as described previously (7).

SNR Studies

To examine the SNR of two-wire catheters, three 25-cmcatheters were constructed using 0.5 mm diameter copperwire insulated with 0.25 mm-thick plastic (wire separationwas 1.6 mm). Each two-wire catheter was terminated in adifferent load impedance: an open circuit, a short circuit,and a 500 pF capacitor. The catheters were then tuned toresonate at 64 MHz, matched to a 50 Ohm line, and de-coupled during RF transmit.

All SNR measurements were made using identical FSEimaging sequences. SNR was calculated in axial images,

FIG. 1. Physical and electrical catheter structure. a: Physical struc-ture of a standard electrophysiology (EP) catheter. b: Unmodifiedelectrical structure of the EP catheter. c: Modified catheter struc-ture. d: Low-frequency topology of the modified catheter. Note thatthe low-frequency structure, which is for intracardiac electrogram(IEGM) recording, is identical to the unmodified catheter structure.e: High-frequency topology of the modified catheter (a loop receiverwith surface electrodes decoupled).

Combined EP/MRI Catheter 595

perpendicular to the catheter shaft. Maximum relativeSNR values were calculated by finding the maximum SNRin each axial image, normalized by the maximum SNRmeasured in any image. All studies were performed on aGE 1.5 T CV/I scanner (General Electric Medical Systems,Milwaukee, WI).

Intracardiac Electrogram Recording

The low-frequency function of the combined functioncatheter was confirmed by measuring the IEGM in a dogduring MR imaging. Images were acquired using a10 frame/sec real-time SSFP sequence (Fiesta, TR �3.2 msec, TE � 1.1 msec, FA � 45°, FOV � 48 � 12cm,ST � 2cm, 128 � 32, BW � 128 kHz) with the cathetersituated in the right ventricle of the dog’s heart. The IEGMwas simultaneously recorded from the catheter tip elec-trodes. This signal was low-pass filtered (cutoff fre-quency � 100 Hz) and passed to a data acquisition systemin the control room through a patch panel, as describedpreviously (7).

All animal protocols were reviewed and approved bythe Animal Care and Use Committee at the Johns HopkinsUniversity School of Medicine. Two mongrel dogs weigh-ing 30–40 kg were anesthetized with a bolus injection ofthiopental and maintained on 1% isoflurane throughoutthe experiment. 8F to 10F introducer sheaths were placedin the right jugular and right femoral veins for catheteraccess and fluid administration.

Catheter Tracking

The high-frequency function of the catheter was demon-strated using the same real-time, 10 frame/sec SSFP se-quence described previously. However, for catheter track-ing the slice select gradient was turned off, yielding pro-jection images.

Simultaneous Catheter Function (EP Recording and MRTracking)

Simultaneous functioning of the catheter was examined bymeasuring the IEGM and concurrently tracking the cathe-ter with a projection imaging sequence. Images were ac-

quired using a 7 frame/sec real-time 2D gradient echosequence (2D FGRE, TR � 4.7 msec, TE � 1.2 msec, FA �20°, FOV � 48 � 12cm, 256 � 32, BW � 64 kHz, no sliceselection). To aid in guidance, a single roadmap image wasobtained using this same pulse sequence with slice selec-tion. In these trials, the catheter was introduced throughthe jugular vein and guided into the right ventricle usingonly MR guidance.

RESULTS

SNR Studies

The SNR of our catheter design was examined by compar-ing it with two other catheter structures: one with an opencircuit and one with a short circuit at the distal end. Theopen circuit catheter is similar to the commercial EP cath-eter design (i.e., there is a high impedance between thedistal lead ends) (Fig. 1b). This catheter produces a verylow SNR, especially near the catheter tip (Fig. 3). Thesecond catheter, with a short circuit at the distal end, is aloop MR receiver. This design is suitable for producing alocal region of high SNR, but cannot receive the IEGM(because the leads are shorted together, bipolar recordingsare not possible). This catheter produces high SNR alongthe entire catheter length (Fig. 3). The SNR is maximalnear the catheter tip.

Our catheter design, with a 500 pF capacitor at the distalend, produced seven times more SNR than the commercial(i.e., open circuit) catheter (Fig. 3) (SNR was compared byaveraging the maximum relative SNR along the catheterlength). The catheter produced 76% of the short circuit(i.e., true loop receiver) SNR. Note that while some SNR islost by using the capacitor rather than the short circuit, weretain the ability to record EP signals with this catheter.

Intracardiac Electrogram Recording

Figure 4 shows a series of long axis slices through a canineheart along with the simultaneously recorded IEGM strip.The arrows indicate the time at which each image wasacquired. Note that the QRS complex, indicating ventric-ular depolarization, precedes contraction of the ventricles.

FIG. 2. Electrical connections of the catheter. a: The 100-cm modified EP catheter is interfaced with the tuning, matching, andsignal-splitting circuit. The intracardiac electrogram (IEGM) and MR signal are connected to their respective receivers. b: Electrical diagramof the catheter and related circuits. The catheter signal is passed through a tuning/matching network to the MR receiver. The IEGM signalis passed to its receiver through RF filters (parallel LC circuits tuned to 64 MHz).

596 Susil et al.

For this series the catheter was positioned in the rightventricle. A bright spot of signal is visible in the rightheart, just proximal to the capacitor between the two elec-trical leads. The tip of the catheter is also visible, movingin and out of the imaging plane as the heart contracts.

Catheter Tracking

Next, the high-frequency function of the catheter was dem-onstrated by receiving the MR signal for device localiza-tion and tracking. This series (Fig. 5) shows that the fulllength of the catheter is well visualized, even in a highlydynamic situation where the shaft flexes and bends afterthe catheter tip becomes stuck. Selected images of thecatheter from a 10-sec push are overlaid on a static, slice-selective roadmap image.

The sequence begins with the catheter tip in the rightventricle (Fig. 5a). The catheter is pulled out several cen-timeters (Fig. 5b) and the tip stays in the right atrium as thecatheter is pushed back in (Fig. 5c), causing the catheterbody to flex. The catheter is withdrawn further (Fig. 5d,e)and the tip again stays in the atrium as the catheter isadvanced (Fig. 5f–h), resulting in more flexing of the shaft.Note that the full length of the catheter is visible, clearlyindicating the position and motion of the catheter as itbends and flexes. We are easily able to appreciate that thecatheter tip is stuck and, therefore, should be withdrawnfurther before attempting to advance back into the ventri-cle.

Simultaneous Catheter Function (EP Recording and MRTracking)

The simultaneous functioning of the catheter was demon-strated by measuring the IEGM and simultaneously track-ing the catheter with a projection imaging sequence. Fig-ure 6 shows selected images, along with concurrently re-corded IEGMs, from a 15-sec catheter push. The cathetersignal is displayed over a static, slice-selective roadmapimage.

FIG. 3. Maximum relative SNR as a function of distance to thecatheter tip for three different two-wire catheters. SNR was calcu-lated in axial images, perpendicular to the catheter shaft. Maximumrelative SNR values were calculated by finding the maximum SNR ineach axial image, normalized by the maximum SNR measured inany image. The open circuit catheter (i.e., the commercial EP cath-eter structure) produced very low SNR. The short-circuited catheter(i.e., an ideal loop receiver) produced the highest SNR. While someSNR is lost by using the capacitive load instead of the short circuit,this catheter can still be used to record EP signals (i.e., its originalfunction).

FIG. 4. Low-frequency catheter function:IEGM recordings during MR imaging. Longaxis images of the canine heart were ac-quired with a 10 frame/sec real-time SSFPsequence. Annotations indicate the left ven-tricle (LV), right ventricle (RV), the position ofthe capacitor, and the electrode tip (Elec.tip). The black arrows indicate the time atwhich each image was acquired. Note thatthe QRS complex, indicating ventricular de-polarization, precedes contraction of theventricles.

Combined EP/MRI Catheter 597

The catheter begins in the jugular vein and is then ad-vanced down the superior vena cava toward the heart (Fig.6a,b). Once the catheter arrives in the right atrium, thecatheter tip gets stuck and the catheter body begins to flex(Fig. 6c). The catheter is withdrawn several centimeters,the shaft is torqued, and then advanced again (Fig. 6d,e).As the tip is now angled anteriorly, it slips through theright atrium and into the right ventricle (Fig. 6f,g). Notethat once the catheter arrives in the ventricle, a largebipolar spike is seen in the intracardiac electrogram. Bi-polar endocardial recordings typically show only electri-cal activation of the myocardium (in the ventricle thesespikes are concurrent with the QRS complex in the bodysurface ECG). In Fig. 6h, the catheter is stably positioned inthe ventricle.

DISCUSSION

Multifunctional Catheter Design

Interventional devices for use in MR-guided and moni-tored procedures must perform several functions. At themost basic level, devices must simultaneously be trackableand able to perform their therapeutic function. More elab-orate devices may perform other functions as well, such asrecording biopotentials from inside the body. Includingseparate hardware to perform each task individually canmake devices prohibitively large and complex. Includingfrequency-dependent elements allows one physical struc-ture to have multiple electrical topologies and functions.

This approach can be used to greatly simplify the con-struction and profile of interventional devices.

Typically, methods for device localization are character-ized as passive or active (12). In passive techniques, in-strument materials are chosen such that the device is vis-ible in the image itself. While inherently simple, theseapproaches are typically plagued by low device contrast,leading to difficult image segmentation. In contrast, activetechniques are based on hardware modifications and post-processing to determine instrument location. While activetechniques typically provide robust signal and easy localiza-tion, they are subject to cumbersome device modifications.

The current approach is attractive in that it extractspositive qualities from both of these techniques. First,similar to passive techniques, the device is visible in theimage itself; no postprocessing is required. In addition,minimal device modification is required. Second, similarto active techniques, the technique produces high contrast,allowing for easy and rapid device localization (Fig. 5).

To visualize the full length of interventional instru-ments, RF receivers that span the entire device are desir-able. This is most easily accomplished using loopless re-ceiver designs (11). Catheters with looped receiver struc-tures, while they tend to be more complicated, producenarrower instrument profiles and easier localization bylimiting signal penetration depth (9,10,13). Because car-diac ablation catheters have two (or more) electrical leads,the catheter is localized by effectively using the secondapproach: a looped receiver.

FIG. 5. High-frequency catheter function: catheter visualization and tracking. Selected images from a 10-sec catheter push acquired witha 10 frame/sec, real-time SSFP sequence. Projection images of the catheter are overlaid on a static, slice-selective roadmap image. Leftventricle (LV), right ventricle (RV), chest wall, and the superior vena cava (SVC). The catheter starts in the right ventricle and is then pulledup into the right atrium (a,b). As the catheter is subsequently pushed, the tip stays in the atrium and the catheter body flexes (c). Thecatheter is pulled back further (d,e). In f–g, the catheter is pushed once again, the tip stays in the right atrium, and the catheter shaft flexes.Note that the full length of the catheter can be easily visualized.

598 Susil et al.

Catheter SNR

By exploiting frequency difference between the MR andintracardiac electrogram (IEGM) signals, a loop receiverwas created while maintaining the original function of theEP catheter. A 500 pF capacitor at the distal end of thecatheter accomplishes this task (Fig. 1). At 64 MHz, thecapacitor has an impedance of only �5j Ohms (nearly ashort circuit), producing a loop receiver. At 60 Hz, wherethe EP signal is recorded, the capacitor has an impedanceof �5j MOhms and the leads are practically isolated.Therefore, the functionality of the EP catheter is main-tained while simultaneously producing a structure withhigh SNR at 64 MHz. While the capacitive load is not anideal short circuit, the SNR with this load approaches thatof the ideal, but impractical, short circuit load.

Device Safety

The safe use of metallic interventional devices during MRimaging is a significant concern (14). There are many fac-tors that affect the heating potential of an interventionaldevice. These include device length, diameter, and theamount of dielectric insulation around the device. We arecurrently characterizing and quantifying the impact ofeach of these factors (15).

Most wire devices, even those that are insulated, havethe potential to cause burns if improperly handled in the

MR scanner (16,17). Care must always be taken to avoidforming loops that may result in high induced currentsand heating. However, compared to other instruments,such as guidewires or needles, EP catheters have the ad-vantage of being heavily insulated—which tends to greatlyreduce unwanted heating.

While the vast majority of the conductor is insulated, thewire tips are exposed at the catheter electrodes. To reducethe potential for heating at these locations we have pro-posed the use of RF tip chokes (Fig. 1). These chokes aretuned to the imaging frequency (64 MHz) and selectivelydecouple the electrodes from the rest of the circuit at thisfrequency (where heating can occur). Therefore, the effec-tive ends of the wire leads are buried within the catheter,proximal to the chokes, and heating potential is reduced.At other frequencies, where the electrodes are needed torecord the IEGM (�100 Hz) and to ablate tissue (kHz), theelectrodes remain connected. We are investigating the ap-plication and efficacy of RF safety chokes for this and otherapplications.

MR-Guided Cardiac Ablation

To perform MR-guided cardiac ablation procedures, cath-eters that are visible under MRI, can receive the IEGM, andcan deliver RF (kHz) energy for tissue ablation are needed.In this study, we have addressed the first two of these

FIG. 6. Simultaneous catheter function: IEGM recordings concurrent with catheter tracking. Selected images from a 15-sec catheter pushacquired with a 7 frame/sec, real-time gradient echo sequence. Left ventricle (LV), right ventricle (RV), chest wall, and the superior vena cava(SVC). The catheter is initially advanced from the jugular vein, down the superior vena cava, toward the right heart (a,b). In c, the cathetertip hits the right atrial wall; then it is withdrawn several centimeters (d) and torqued. The catheter is advanced again (e), passes through theright atrium (f), and into the right ventricle (g). In h, the catheter is seated in the right ventricle. Note that a strong bipolar signal is recordedonce the catheter tip arrives in the right ventricle (lower amplitude signal is also seen from the right atrium).

Combined EP/MRI Catheter 599

requirements. In previous work by our group the ability toperform RF ablation in the MR scanner was demonstrated(7). Also, it was shown that ablation lesions could bedetected 2 min after RF delivery using T2-weighted spinecho and contrast enhanced T1-weighted gradient echosequences. Further work is needed to demonstrate RF ab-lation combined with active tracking of the EP catheter.While high Q filters must be added to prevent ablationenergy from damaging the MR preamplifiers, no majorchanges to the present catheter design are required.

A further issue to consider is active visualization of thecatheter tip. With the current design, tip electrodes of thecatheter do not generate high signal. In fact, if the electrodetips are fully choked from the rest of the catheter circuit toreduce tip heating (Fig. 1), there should be little signalreceived here. The highest signal is received near the tipcapacitor (Fig. 3), which is approximately 2 cm from theend of the catheter. We are exploring new designs that willallow for improved tracking of the catheter tip.

CONCLUSION

In the present study, a combined EP/MR imaging catheterwas designed and implemented. Using a relatively simplehardware structure, the intracardiac electrogram was re-corded while the full length of the catheter was activelytracked using real-time MR imaging sequences. The cath-eter behaves simultaneously as a two-lead electrophysiol-ogy catheter (for recording the intracardiac electrogram)and as a loop receiver for catheter visualization. This de-sign will allow us to capitalize on the excellent tissuecontrast and functional imaging capabilities of MRI toimprove the efficiency and efficacy of cardiac EP interven-tions.

ACKNOWLEDGMENTS

The authors thank Ritsushi Kato, Ananda Kumar, YusufOner, Ronald Ouwerkerk, and Andrew Yung for assistancewith the experiments and Mary McAllister for editorialassistance. Robert Susil is supported by an NIH traininggrant, Christopher Yeung is supported by the WhitakerFoundation and an NIH training grant, and Dr. Halperin is

a McClure Fellow of the Johns Hopkins Applied PhysicsLaboratory.

REFERENCES1. Calkins H. Catheter ablation for cardiac arrhythmias. Med Clin N Am

2001;85:473–502, xii.2. Feld GK, Fleck RP, Chen PS, Boyce K, Bahnson TD, Stein JB, Calisi CM,

Ibarra M. Radiofrequency catheter ablation for the treatment of humantype 1 atrial flutter. Identification of a critical zone in the reentrantcircuit by endocardial mapping techniques. Circulation 1992;86:1233–1240.

3. Jackman WM, Beckman KJ, McClelland JH, Wang X, Friday KJ, RomanCA, Moulton KP, Twidale N, Hazlitt HA, Prior MI, et al. Treatment ofsupraventricular tachycardia due to atrioventricular nodal reentry, byradiofrequency catheter ablation of slow-pathway conduction. N EnglJ Med 1992;327:313–38.

4. Kay GN, Epstein AE, Dailey SM, Plumb VJ. Role of radiofrequencyablation in the management of supraventricular arrhythmias: experi-ence in 760 consecutive patients. J Cardiovasc Electrophysiol 1993;4:371–389.

5. Saliba WI, Natale A. Ventricular tachycardia syndromes. Med Clin NAm 2001;85:267–304.

6. Singer I. Interventional electrophysiology. Baltimore: Williams andWilkins; 1997.

7. Lardo AC, McVeigh ER, Jumrussirikul P, Berger RD, Calkins H, Lima J,Halperin HR. Visualization and temporal/spatial characterization ofcardiac radiofrequency ablation lesions using magnetic resonance im-aging. Circulation 2000;102:698–705.

8. Atalar E, Bottomley PA, Ocali O, Correia LC, Kelemen MD, Lima JA,Zerhouni EA. High resolution intravascular MRI and MRS by using acatheter receiver coil. Magn Reson Med 1996;36:596–605.

9. Burl M, Coutts GA, Herlihy DJ, Hill-Cottingham R, Eastham JF, HajnalJV, Young IR. Twisted-pair RF coil suitable for locating the track of acatheter. Magn Reson Med 1999;41:636–638.

10. Ladd ME, Erhart P, Debatin JF, Hofmann E, Boesiger P, von SchulthessGK, McKinnon GC. Guidewire antennas for MR fluoroscopy. MagnReson Med 1997;37:891–897.

11. Ocali O, Atalar E. Intravascular magnetic resonance imaging using aloopless catheter antenna. Magn Reson Med 1997;37:112–118.

12. Ladd ME, Quick HH, Debatin JF. Interventional MRA and intravascularimaging. J Magn Reson Imag 2000;12:534–546.

13. Ladd ME, Zimmermann GG, Quick HH, Debatin JF, Boesiger P, vonSchulthess GK, McKinnon GC. Active MR visualization of a vascularguidewire in vivo. J Magn Reson Imag 1998;8:220–225.

14. Ordidge RJ, Shellock FG, Kanal E. A Y2000 update of current safetyissues related to MRI. J Magn Reson Imag 2000;12:1.

15. Yeung CJ, Susil RC, Atalar E. RF safety of wires in interventional MRI:using a safety index. Magn Reson Med 2001;47:187–193.

16. Bashein G, Syrory G. Burns associated with pulse oximetry duringmagnetic resonance imaging. Anesthesiology 1991;75:382–383.

17. Shellock FG, Slimp GL. Severe burn of the finger caused by using apulse oximeter during MR imaging. AJR Am J Roentgenol 1989;153:1105.

600 Susil et al.