How to Perform Rotor Mapping for Atrial Fibrillation Ablation

25


How to Perform Rotor Mapping for Atrial Fibrillation Ablation


Sanjiv M. Narayan, MD,1 PhD; Mohan Vishwanathan, MD; Christopher A. B. Kowalewski; Tina Baykaner, MD; David E. Krummen, MD; Paul J. Wang, MD


Introduction


Ablation is increasingly used to treat atrial fibrillation (AF), yet while pulmonary vein isolation (PVI) helps many patients, its 1- to 2-year success rate with contemporary catheters is 40%–50% for persistent AF1,2 and 50%–65% for paroxysmal AF.35 Attempts have been made to improve PVI by ablating linear lesions or empiric electrogram targets, yet these often extensive procedures have not improved results versus PVI alone in multicenter trials.1,2,6


Ablation targeting AF sources has gained much attention in recent years. The source model states that AF is sustained by rotational (rotor) or focal sources arising at patient-specific locations in either atria that cause disordered activity. This has been shown by human optical mapping of AF,7 clinical AF rotor ablation,812 optically mapped canine13 and sheep14 AF, mathematical and in silico analyses.15 This model can reconcile the clinical paradox that limited ablation terminates AF in some patients,9,11,16 despite its peripheral disorder, yet extensive (untargeted) ablation can be ineffective in others.1,2,17


These observations contradict the multiwavelet hypothesis, in which AF is self-sustaining disorder with no localized drivers.18,19 That model predicts that more extensive ablation should increase success by limiting critical mass for disordered waves, yet linear lesion sets as well as extensive procedures have not improved results in multicenter trials1,2,6 and even extensive Maze surgery has suboptimal results in contemporary studies20 compared to original reports.


Rotor ablation is a novel clinical procedure with promising results at many centers,8,10,12,21,22 by multiple techniques.9,2325 This practical chapter focuses on focal impulse and rotor modulation (FIRM), for which many clinical studies have now been reported. While most have shown promising results, a few have been disappointing. We carefully analyze both sets of studies to determine the extent to which they can be reconciled by methodological, population, or other factors. We focus on the mechanistic rationale for each procedural component, which is the foundation upon which AF rotor and source ablation have been developed.


Clinical Basis of FIRM-Guided Rotor Ablation


FIRM-guided ablation is based on the concept that AF results from triggers and substrates. Triggers at PV and non-PV sites have been shown to actually initiate AF by uncovering rate-related conduction slowing, producing conduction block, spiral wave reentry and AF.26,27 FIRM-guided ablation and PVI are often used in patients with persistent and long-standing persistent AF. FIRM is also used in paroxysmal AF patients with sustained AF after PVI, who often received substrate ablation in addition to PVI in seminal trials of paroxysmal AF,2830 and in patients with recurrent paroxysmal AF after previously failed PVI, in whom non-PV substrates have been reported.31


Table 25.1 summarizes results of FIRM-guided ablation after the original CONFIRM trial9 and multicenter FIRM registry.8 There is heterogeneity in the results of FIRM-guided ablation for AF across centers, but this does not appear to stratify by clinical populations. For instance, Sommer et al.21 reported 80% single-procedure freedom from AF and atrial tachycardia by FIRM ablation (including FIRM-only cases) in challenging patients with multiple prior failed procedures and long-standing persistent AF; and Miller,8,22 Tomassoni,12 Rashid,10 and others demonstrated 70%–80% single-procedure success in broad populations including similar patients and those with paroxysmal AF; yet Buch et al.32 reported 21% success in patients, half of whom had paroxysmal AF, despite also performing PVI (most with several prior ablations).



Discrepancies between centers thus likely reflect differences in technique rather than patient type. Lower success from rotor ablation was seen at centers with fewer cases per operator and in early case studies (i.e., pre-2013) before software had automatic detection of rotor targets and relied more upon operator experience (Table 25.1). Follow-up duration is often a confounder, yet in the case of FIRM-guided ablation, promising single-procedure success is reported at follow-up of 1 to 2 years,22 2 years,12 and 3 years33 by separate groups.



image


Figure 25.1 Human atrial fibrillation is sustained by stable rotational sources in human atria on optical mapping.7,34 Panel A: Rotational AF source in RA, stable on endocardium yet transient/meandering on epicardium. Panel B: Ablation at AF source terminates AF to sinus rhythm or atrial tachycardia.7 Panel C: AF sources in left atria near and remote from PVs, anchored to microfibrosis and fiber angles that are more complex on endocardium than epicardium. Panel D: Electrograms in AF comprise far-field signals and poorly represent local activity (optical action potentials).34


This chapter is designed to systematically cover technical aspects of the procedure in terms of basket placement, map interpretation to identify rotor targets (i.e., incorporating data from both early studies without and contemporary studies with automatic identification), the rotor ablation procedure itself, and procedural endpoints.


Mapping the Atria with Basket Catheters


A core principle of FIRM-guided ablation is to map the right and left atria widely using multipolar catheters to analyze many wavefronts at the same time rather than sequentially. The need for wide mapping is indicated by the abundance of bench-to-bedside studies showing that the small number of rotor and focal sources that sustain AF may arise essentially anywhere within either atria.


Figure 25.1 shows results from the gold standard of optical mapping applied to human AF,7,34 showing rotational AF drivers in the right atrium (RA; Figure 25.1A) where ablation terminated AF to sinus rhythm (Figure 25.1B), and two concurrent AF sources in left atrium (LA; Figure 25.1C). Sources were stabilized by microfibrosis present in patients with comorbidities that may be absent in animals. Note that while action potentials are fairly regular in Figure 25.1D, electrograms show far-field activations that will alter contour maps of electrogram timings35,36 and render them complex. This may explain differences between optical maps and FIRM maps, which show similar AF rotor characteristics in simultaneous optical/FIRM studies of human atria,37 and traditional electrogram-derived AF contour maps. Indeed, many studies show that clinical AF electrograms show far-field artifact that often poorly reflects local activity.38 FIRM algorithms were derived from monophasic action potentials,38,39 conduction velocity,26,27 and signal processing40,41 to improve the accuracy of AF maps.42


Figure 25.2 summarizes two common approaches to FIRM-guided cases. Surveying the RA is crucial, since 30% or more of AF sources lie in RA9,24 as also shown classically.43 Mapping the RA first is designed to identify sources before ablation terminates AF, when sources may no longer be inducible. A variant (Figure 25.2B) is to map RA sources then LA sources, then ablate sources in either order in combination with PVI. Repeat mapping is strongly recommended in any workflow to verify AF source elimination.



image


Figure 25.2 Clinical approaches to FIRM-guided rotor mapping. Panel A: Traditional workflow. Panel B: Alternative workflow.


We have previously described basket positioning in detail.44 Figure 25.3 shows optimal and suboptimal basket placement. Multiple basket sizes are available (48- and 60-mm diameter by Boston Scientific; 50-, 60-, and 70-mm diameter by Abbott Electrophysiology) and are selected from CT or echocardiographic atrial dimensions. Figure 25.3A shows a neutral LA basket position, where even this large basket does not contact the atrial walls in this large atrium (see its spherical shape). We thus repositioned this basket like any mapping catheter to map most of the atria in 2–3 epochs (1–3 minutes each), with spline deformation indicating good contact (Figure 25.3B). Suboptimal basket position, as illustrated45 in Figure 25.3C, may explain suboptimal outcomes from ablation, where the basket is free floating (spherical), far from the coronary sinus (LA floor) and left main bronchus (LA roof).



image


Figure 25.3 Pearls and pitfalls in basket positioning for FIRM mapping.
Panel A: Neutral basket position in LA, which is suboptimal in this large atrium.
Panel B: Multi-position basket mapping of large atrium, which is commonly done now that remapping is rapid, with contact indicated by deformed splines that are no longer spherical. The basket was torqued clockwise to cover right PV antra, and advanced with counterclockwise torque for anterolateral LA.
Panel C: Suboptimal free floating baskets, indicated by spherical shape and large distance to coronary sinus and roof.


Theoretically, baskets provide sufficient spatial resolution to identify 1- to 2-cm diameter reentry circuits in human AF, as predicted by Allessie et al.47 and ourselves48 and confirmed in human optical maps (Figure 25.1A).7 It is important to recognize that there is a spatial resolution, which is too high for AF mapping given precision in marking electrograms (i.e., why a microscope is not used to measure macroscopic distances). If electrodes are too closely spaced, wavefronts will arrive so rapidly at successive sites that differences in timing may fall within measurement errors. The precision in marking often complex AF electrograms is ~5 to 10 ms. If atrial conduction velocity is reduced at 40 cm/sec in AF patients,26,27 an electrode spacing of 0.2 to 0.4 cm (= 40 × 0.005 or 40 × 0.010) is the physiologically minimum separation. Mapping with higher resolution35,36,49 may define circuits falling within the errors of measurement.


Basket mapping is thus simple in principle, but specific considerations are required to optimize AF mapping. Our estimate from the published rotor mapping studies and from experience is that this likely requires > 10 cases per operator for optimal rotor mapping and ablation. These estimates may be modified with additional advances in basket design and algorithmic interpretation of AF activity.


Interpreting Spatial Maps of AF Activity


Interpreting AF maps to identify rotational or focal sources in the midst of fibrillatory disorganization is critical for FIRM-guided ablation, and novel to most proceduralists.


FIRM portrays AF as movies in an atrial grid orientation (Figure 25.4). Rotational activity is identified by grid coordinates of the core (“pivot of rotation”) over time, typically over tens of seconds (i.e., dozens of cycles), in the midst of disorder. Figure 25.4A indicates clockwise rotation around a rotor core indicated by the red rotational activity profile (RAP) region of calculated phase singularities for hundreds of cycles (i.e., the entire minute of recording). Figure 25.4B shows two concurrent LA sources in a patient with AF, similar to those imaged optically in human AF (Figure 25.1C).34 Focal sites show repetitive activity emanating from an origin, also with peripheral disorder. Endocardial AF sources precess but are stable in ~2 cm2 areas (1 to 2 electrodes per axis). It is interesting that epicardial mapping by ECGi shows less-stable AF sources in larger areas,24 still conserved over time, that may reflect instability of epicardial manifestations of endocardial sources on optical imaging of human atria,7 or technical issues with the inverse solution.50



image


Figure 25.4 Successive FIRM maps from a movie of rotational sources in AF. Panel A: Clockwise rotation around rotor core (RAP) computed algorithmically over hundreds of cycles. One cycle length shown (185 ms). Panel B: Two concurrent rotational AF sources, clockwise (top) and counterclockwise (bottom) of slightly different cycle lengths. Overall map duration is approximately 1 cycle of both sources (duration 174 ms). Ablation targets the RAP regions of about 2 to 3 cm2 based on electrode coordinates on the atrium (see Figure 25.5).

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Aug 27, 2018 | Posted by in CARDIOLOGY | Comments Off on How to Perform Rotor Mapping for Atrial Fibrillation Ablation

Full access? Get Clinical Tree

Get Clinical Tree app for offline access