Case Study

A Robot Can Do Better: Perspectives During a Case of Atrial Tachycardia

Aseem Desai, MD, FHRS

Co-Director, Mission Heritage Heart Rhythm Specialists; Mission Viejo, California

Aseem Desai, MD, FHRS

Co-Director, Mission Heritage Heart Rhythm Specialists; Mission Viejo, California

Case Description

A 73-year-old male with a history of paroxysmal atrial fibrillation (PAF) and atrial flutter presented with recurrent episodes of narrow complex tachycardia. In the emergency room, the patient was administered adenosine, which revealed morphology and cycle length consistent with atrial tachycardia (AT). Diltiazem failed to convert the rhythm, but electrical cardioversion was successful. Given that the patient was highly symptomatic with near-syncope and the tachycardia was recurring despite medical therapy, he was taken to the electrophysiology lab for mapping and possible ablation.

The patient’s arrhythmia history dates back to 2009 when he underwent a pulmonary vein isolation (PVI) and left atrial (LA) roof line ablation procedure for recurrent symptomatic PAF failing drug therapy. AF recurred in 2012, and three-dimensional (3D) electroanatomic mapping revealed reconnection in three out of four pulmonary veins (CARTO, Biosense Webster, Inc., a Johnson & Johnson company). The patient underwent redo PVI, reinforcement of the LA roof line, and empiric right atrial (RA) cavotricuspid isthmus ablation. The patient did well until six months ago when he began having the episodes of narrow complex tachycardia referenced above.  

The medical history is otherwise unremarkable, with no other significant risk factors or triggers for atrial arrhythmias such as sleep apnea, thyroid disease, obesity, hypertension, diabetes, or heavy alcohol use. Metoprolol maintained normal sinus rhythm after the 2012 ablation.

At the start of the current ablation procedure, the patient was in clinical tachycardia (Figure 1). Three-dimensional activation and voltage mapping of the LA and RA was performed (PENTARAY Catheter, Biosense Webster, Inc.). Extensive scarring was noted in both atria, indicating multiple potential substrates for arrhythmias. Activation mapping was suggestive of a focal tachycardia, and the demonstration of concealed entrainment suggested a reentry mechanism (micro-reentry). In the LA, there was no discrete area of early activation relative to the reference catheter in the coronary sinus. The earliest region was along the anterior portion of the interatrial septum. Earliest activation in the RA was seen near the crista terminalis (CT) along the mid-posterior wall (Figure 2).

To better understand the propagation of the AT, a Ripple Map (Biosense Webster, Inc.) was created and superimposed on the activation and bipolar voltage maps. These maps were compared side by side to understand the relationship between arrhythmia movement and substrate. Ripple Mapping is not dependent on local point annotation or Window of Interest settings. This type of map integrates activation and bipolar voltage data into one propagation. By visualizing local bipolar voltage/signal characteristics as the wavefront propagates, the identification of multicomponent signals and associated activation patterns can be better identified.1 The Ripple Map demonstrated origin of the AT from the mid-posterior RA with breakthrough across the septum to the LA. The voltage at the origin was a border-zone region between healthy and diseased tissue (Video 1).

When high-density mapping was performed in the earliest area of the RA in the mid-posterior wall, an activation propagation map confirmed the mechanism as focal (Video 2).

When high-density mapping was performed in the earliest area along the inferior-posterior RA, a long, high-frequency, low-amplitude, complex fractionated electrogram signal was identified. The duration of the signal made up almost 50 percent of the tachycardia cycle length (Figure 3).

Due to the known proximity of the phrenic nerve, the structure was identified and tagged with high-output pacing (20 Volts, 2.0 milliseconds). Manual palpation of the diaphragm was used to assess capture. Almost the entire region of early activation had phrenic nerve capture except for small isolated areas. There was no diaphragmatic capture at these sites.

A remote magnetic navigation (RMN) mapping and ablation catheter (NAVISTAR RMT THERMOCOOL Catheter, Biosense Webster, Inc.) was advanced through a long steerable sheath (MOBICATH Sheath, Biosense Webster, Inc.) to the area of interest using the Stereotaxis Niobe Robotic Magnetic Navigation System (Stereotaxis, Inc.). Ablation was performed with 25 watts with immediate termination of the tachycardia, rendering it non-inducible (Figure 4).


There are several points worth discussing in more detail. Focal atrial tachycardia (FAT) is the least common type of supraventricular tachycardia, accounting for 5-15% of cases presenting to the electrophysiology lab for ablation. This is in comparison to AV nodal reentry and accessory pathways. Mechanisms of FAT include abnormal automaticity, triggered activity, and micro-reentry. The arrhythmia commonly originates from specific anatomic sites due to anatomic and electrical heterogeneity in depolarization and repolarization. In published series, the right atrium is the most common location, accounting for about 75% of cases. Of these, approximately 33% originate from the CT, particularly the superior and mid portions. Other frequent sites include the tricuspid annulus, the coronary sinus ostium and body, the perinodal (parahisian) region and septum, and from within the right atrial appendage. In the left atrium, the majority of foci originate from the pulmonary veins, mitral annulus, left atrial appendage, and less commonly, the left septum.2 

Although this patient’s atrial tachycardia focus was not at the CT, it was near it and some of the same concerns apply about phrenic nerve injury. For this reason, it is worth discussing further. Of right atrial tachycardias (RAT), about two-thirds occur in the absence of structural heart disease and are distributed along the long axis of the CT.3 Most of these are in the superior and mid portions of the CT. Regarding its arrhythmogenic properties, the CT is an area of slow transverse and rapid linear conduction, setting up the substrate for re-entry. Ablation of CT tachycardias carries a small risk of damage to the right phrenic nerve. To avoid this complication, transient high-output pacing prior to ablation to confirm no diaphragmatic capture is usually done. If there is phrenic nerve stimulation in a desired ablation target, placing a pericardial balloon via a subxiphoid approach to distance the nerve from the ablation site has been successfully used.4 Cryoablation is another technique utilized in these cases.5 

In this particular case, RMN was extremely helpful in providing catheter stability and maneuverability. The system is composed of three components: two focused-field permanent magnets, an irrigated magnetic catheter, and a catheter advancing/retracting system (Cardiodrive, QuikCAS; Stereotaxis, Inc.). The two magnets emit a 0.08 to 0.1 Tesla magnetic field and are located on either side of the patient’s body. The magnetic catheter has three embedded magnets and can be actively deflected by changing the magnetic field orientations using a computer mouse or keypad. The irrigated magnet catheter can be steered leftward or rightward following the direction of the vectors by a magnet field-controlled system, and can be advanced or retracted as short as 1 millimeter with the Cardiodrive unit.



Because external magnetic fields cause the catheter to “stick” to cardiac tissue, and because the catheter is soft and floppy, it remains in continuous contact with tissue despite a beating heart and breathing lungs. Furthermore, because up to 1 mm in precision accuracy can be achieved with catheter movement, this type of technology is especially useful for ablating near critical anatomic structures. Several studies have been published regarding the safety and efficacy of RMN ablation for atrial and ventricular arrhythmias. Furthermore, studies comparing manual versus RMN have demonstrated similar efficacy.6,7



Another point to make in this case is what the ideal pacing output should be when checking for phrenic capture. To date, there have been no conclusive studies indicating what the ideal output should be.8 We did notice that there were areas that did not capture the phrenic at 10V, but did at 20V. Certainly more tissue is recruited with the larger output, but it raises the question of where to draw the line between “undertreating due to over-capture” and “overtreating due to under-capture.”

There are many limitations with identifying the course of the phrenic nerve using high-output pacing. Deep respiration, cardiac motion, and patient motion all contribute to catheter instability, and therefore, reliability of the exact course of the phrenic nerve. With RMN, many of these issues are not present due to the continuous catheter contact that is achieved. 

Disclosures: Dr. Desai has no conflicts of interest to report regarding the content herein.

For more information about Dr. Desai, please visit Desai can be reached on Twitter at @DrAseemDesai. He is also available on Facebook, InstagramYouTube, and LinkedIn.

Exclusive Online Content: 

Watch Dr. Desai provide a video commentary of this article:

Don't miss the bonus quiz also available for this article! 

Bonus video shows an example of RMN catheter vector movement (not related to this case):  

  1. CARTO CONFIDENSE Module with Ripple Mapping. Biosense Webster. Available at 
  2. Feldman A, Kalman JM. Electrocardiogram recognition and ablation of atrial tachycardia. European Cardiology. 2010;6(4):58-63.
  3. Kalman JM, Olgin JE, Karch MR, Hamdan M, Lee RJ, Lesh MD. ‘Cristal tachycardias’: origin of right atrial tachycardias from the crista terminalis identified by intracardiac echocardiography. J Am Coll Cardiol. 1998;31:451-459.
  4. Saoudi N, Cosio F, Waldo A, et al. Classification of atrial flutter and regular atrial tachycardia according to electrophysiologic mechanism and anatomic bases: a statement from a joint expert group from the Working Group of Arrhythmias of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology. J Cardiovasc Electrophysiol. 2001;12:852-866.
  5. Bastani H, Insulander P, Schwieler J, et al. Safety and efficacy of cryoablation of atrial tachycardia with a high risk of ablation-related injuries. Europace. 2009;11:625-629.
  6. Chun KR, Wissner E, Koektuerk B, et al. Remote-controlled magnetic pulmonary vein isolation using a new irrigated-tip catheter in patients with atrial fibrillation. Circ Arrhythm Electrophysiol. 2010;3:458-464.
  7. Jia K, Jin Q, Liu A, et al. Remote magnetic navigation versus manual control navigation for atrial fibrillation ablation: a systematic review and meta-analysis. J Electrocardiol. 2019;55:78-86.
  8. Mears J, Lachman N, Christensen K, Asirvatham SJ. The phrenic nerve and atrial fibrillation ablation procedures. J Atr Fibrillation. 2009;2:176.