Author + information
- Received July 24, 2017
- Revision received December 8, 2017
- Accepted December 11, 2017
- Published online April 16, 2018.
- Anthony Aizer, MD, MSc∗ (, )
- Austin V. Cheng, BS,
- Patrick B. Wu, MD,
- Jessica K. Qiu, BS,
- Chirag R. Barbhaiya, MD,
- Steven J. Fowler, MD,
- Scott A. Bernstein, MD,
- David S. Park, MD,
- Douglas S. Holmes, MD and
- Larry A. Chinitz, MD
- New York University Cardiac Electrophysiology Service, New York University School of Medicine, New York University Langone Medical Center, New York, New York
- ↵∗Address for correspondence:
Dr. Anthony Aizer, NYU Heart Rhythm Center, New York University Langone Medical Center, Tisch Hospital, Electrophysiology Laboratory, 560 First Avenue, 5th Floor, New York, New York 10016.
Objectives This study sought to investigate the effect of pacing mediated heart rate modulation on catheter–tissue contact and impedance reduction during radiofrequency ablation in human atria during atrial fibrillation (AF) ablation.
Background In AF ablation, improved catheter–tissue contact enhances lesion quality and acute pulmonary vein isolation rates. Previous studies demonstrate that catheter–tissue contact varies with ventricular contraction. The authors investigated the impact of modulating heart rate on the consistency of catheter–tissue contact and its effect on lesion quality.
Methods Twenty patients undergoing paroxysmal AF ablation received ablation lesions at 15 pre-specified locations (12 left atria, 3 right atria). Patients were assigned randomly to undergo rapid atrial pacing for either the first half or the second half of each lesion. Contact force and ablation data with and without pacing were compared for each of the 300 ablation lesions.
Results Compared with lesion delivery without pacing, pacing resulted in reduced contact force variability, as measured by contact force SD, range, maximum, minimum, and time within the pre-specified goal contact force range (p < 0.05). There was no difference in the mean contact force or force–time integral. Reduced contact force variability was associated with a 30% greater decrease in tissue impedance during ablation (p < 0.001).
Conclusions Pacing induced heart rate acceleration reduces catheter–tissue contact variability, increases the probability of achieving pre-specified catheter–tissue contact endpoints, and enhances impedance reduction during ablation. Modulating heart rate to improve catheter–tissue contact offers a new approach to optimize lesion quality in AF ablation. (The Physiological Effects of Pacing on Catheter Ablation Procedures to Treat Atrial Fibrillation [PEP AF]; NCT02766712)
Radiofrequency catheter ablation generates resistive and conductive heating of cardiac tissue to cause cell death and scar formation, with the goal of halting electrical conduction. Critical to this is consistently achieving sufficiently elevated tissue temperatures (1,2). Initially identified factors affecting tissue temperature and lesion formation included quantity of radiofrequency energy delivered over time (wattage) and temperature conduction properties to surrounding tissue. Recently, catheter–tissue contact has been shown to affect lesion size and consistency by modulating how much of the emitted radiofrequency energy from the catheter reaches the tissue (3). With this understanding, catheters that measure and provide real-time data on catheter–tissue contact have been developed with the hopes of improving ablation outcomes.
Subsequently, clinical data have shown that achieving specific catheter–tissue contact goals, as defined by contact force, predicts improved outcomes with ablation of atrial fibrillation (AF) (4). In particular, as the percentage of time spent within a pre-specified contact force range increases, the likelihood of ablation success increases (5). However, there are few data on how to best achieve specific contact force goals. It has also been shown that cardiac motion significantly affects contact force (6,7). We attempted to study how cardiac motion can be modulated to improve catheter–tissue contact. In particular, we sought to determine if cardiac pacing to increase the cardiac ventricular rate would improve catheter–tissue contact by reducing the extremes of contact force during ablation lesions. In addition, we assessed whether this pacing-mediated modification of catheter–tissue contact correlated with improved measures of lesion delivery as reflected by impedance reduction during radiofrequency energy delivery.
This study was approved by the Institutional Review Board of the New York University School of Medicine, in compliance with the Declaration of Helsinki. Twenty patients with paroxysmal AF who presented in sinus rhythm were randomized to 1 of 2 study groups. All patients received general anesthesia with intubation to suppress spontaneous respiration and prevent patient movement. Based on previously published data examining contact force and ventricular contraction, 20 patients were enrolled in this study (8). To standardize the potential effect of respiration on catheter motion, ventilation was performed in all patients with a tidal volume of 500 ml and a respiratory rate of 12 breaths per minute. All antiarrhythmics were held for at least 5 half-lives before the procedure. Group 1 received atrial pacing during the first half of each lesion with sinus rhythm during the second half; in group 2, ablation occurred during sinus rhythm for the first half of the lesion and pacing during the second half. Pacing was performed from a single site at twice diastolic threshold up to 10 milliamps for 2.0 ms, using a 20-pole catheter (Livewire Duo-decapolar, St. Jude Medical, Inc., St. Paul, Minnesota) wrapping around the right atrium and extending to the distal coronary sinus. Before ablation, pacing was performed at 500 ms from a site in which atrial capture was demonstrated to be consistent. Atrial pacing, rather than ventricular pacing, was selected to maintain the hemodynamics of atrioventricular (AV) synchrony and to ensure consistent ventricular timing without the risk of ventricular pacing and superimposed AV conduction. Patients who developed AF during the study were cardioverted to sinus rhythm. If AV block occurred when pacing at 500 ms, the rate was sequentially reduced to 550 ms, and if AV block occurred then 600 ms. If AV block occurred at 600 ms, the pacing catheter was to be placed in the right ventricle and paced at 500 ms.
Using the Carto 3 mapping and ablation system (Biosense Webster, Inc., Irvine, California), all participants received ablation lesions at 15 pre-specified locations (12 left, 3 right atria). Six lesions each were placed around the left and right pulmonary veins at the following locations: superior-posterior, mid posterior, inferior-posterior, superior-anterior, mid anterior, inferior-anterior. As is standard in our electrophysiology laboratory to optimize patient safety and outcome, lesions at the 6 posterior wall locations were limited to 20 s, whereas lesions at the anterior wall locations were limited to 30 s. Three lesions each of 30-s duration were placed at the anterior, mid, and posterior aspects of a typical cavotricuspid isthmus ablation. If the operator noted the ablation catheter to have a dramatic motion indicative of displacement or if catheter motion exceeded 2.5 mm within the magnetic field defined by the mapping system, the lesion was aborted. A repeat lesion was performed after sufficient time was given to allow for tissue cooling, maximally separated from all previous lesions while remaining in the appropriate anatomic location as prescribed by the protocol. All lesions were applied using a THERMOCOOL SMARTTOUCH 3.5-mm ablation catheter (Biosense Webster, Inc.) and a steerable sheath (Agilis NxT, St. Jude Medical). Based on previously published data on typical contact force range goals of other medical centers and the operator contact force goals at our institution, it was pre-specified that the goal contact force for each lesion would be between 10 and 40 g (5). Ablations were applied in a power control mode with immediate application at 30 Watts. Catheter irrigation was applied at 15 ml/h.
Data were extracted from the mapping system and reviewed to assess contact force data. Contact force and impedance was recorded every 50 ms. Data for each lesion were recorded in their entirety, and correlated with both lesion location and the presence or absence of pacing. From each lesion, 4 s of contact force data (2 s from each half of the 4 s midway through the entire lesion) was excluded from the analyses to account for transition between the paced rhythm and sinus rhythm. Each 20- and 30-s lesion resulted in 320 and 520 data points, respectively, with each case comprising 6,600 data points.
The pre-specified primary endpoint of the analysis was contact force variability, defined as the SD of the contact force. Additional endpoints included the range, and minimum and maximum contact force with and without pacing. The percentage of time spent within the pre-specified contact force range also was compared with and without pacing.
To assess for the possibility of operator bias in catheter manipulation, the mean contact force and the force–time integral was compared between pacing and nonpacing time points for each lesion. Finally, to assess if heart rate modulation affected lesion quality, the absolute reduction in impedance during the first 10 s of each lesion was compared between pacing first lesions and pacing second lesions (9).
Means and proportions of baseline clinical characteristics and outcomes between the 2 study arms were compared using Mann–Whitney U tests for continuous variables and Fisher’s exact tests for categorical variables. Correlations of continuous variables were evaluated using Spearman rank correlation coefficients. All hypotheses tests were 2 tailed with p < 0.05 considered significant. Analyses were performed using SPSS software version 21 (International Business Machines Corporation, Armonk, New York).
All 20 patients had paroxysmal AF, while presenting in sinus rhythm on the day of the procedure. There were no differences in baseline clinical characteristics (Table 1). In particular, there was no difference in baseline heart rate, left atrial size, or left ventricular ejection fraction. Nineteen of 20 patients were paced at 500 ms, and 1 patient was paced at 550 ms, secondary to AV block at 500 ms. One patient developed AF during the sinus rhythm portion of an ablation lesion, requiring cardioversion. No adverse events were noted in either randomization arm.
Distribution of contact force
Compared with lesion delivery without pacing, pacing resulted in reduced contact force variability, assessed by multiple parameters (Figure 1). The SD of the contact force (primary endpoint) was significantly lower with pacing when compared with nonpacing during sinus rhythm (Table 2). This reduction in the contact force SD was consistent across all locations in the left and right atria, except the right superior posterior aspect of the left atrium (Figure 2). In addition, the range of the contact force was significantly reduced (p = 0.001) during pacing. The maximum contact force achieved during lesion delivery was significantly lower and there was a trend toward an increase in the minimum contact force. This improved consistency of contact force was also noted by a greater percentage of time within the prescribed contact force range of 10 to 40 g with pacing when compared with sinus rhythm (Table 2). Although contact force variability was reduced during pacing, there was no overall difference in mean contact force or force–time integral between pacing and nonpacing lesions.
Regional variability of mean contact force
Reduced contact force variability with unchanged mean contact force was found consistently across the majority of locations throughout the left and right atria. However, regional analyses demonstrated significant differences in mean contact force between pacing and nonpacing at the left mid posterior and the right superior posterior aspects of the left atrium. Atrial pacing with the catheter at the left mid posterior aspect of the left atrium resulted in a nonsignificant trend toward a lower mean contact force (17.1 ± 5.6 g vs. 18.6 ± 4.8 g; p = 0.058). This was driven by a statistically significant reduction in the maximal contact force with pacing (28.4 ± 9.4 g vs. 31.8 ± 10.0 g; p = 0.041) without any significant change in the minimum contact force per lesion (8.6 ± 6.5 g vs. 8.6 ± 4.9 g; p = 0.98). Atrial pacing with the ablation catheter at the right superior posterior aspect of the left atrium significantly increased the mean contact force (27.6 ± 12.8 g vs. 22.1 ± 47.4 g; p = 0.021). The mean contact force increased by significantly increasing the minimum contact force (13.1 ± 7.4 g vs. 9.8 ± 5.0 g; p = 0.009) without significantly increasing the maximum contact force (53.8 ± 31.0 g vs. 50.4 ± 22.9 g; p = 0.45) (Table 3).
Effect of pacing on impedance changes during ablation
To determine the potential effects of pacing on lesion quality, the impedance reduction during the first 10 s of each pacing first lesion was compared with the impedance reduction of the first 10 s of each nonpacing first lesion. Pacing resulted in a statistically significant 30% greater impedance reduction at 10 s when compared with nonpaced lesions (15.7 ± 8.3 Ω reduction vs. 12.2 ± 7.4 Ω reduction; p < 0.001). Regional analyses demonstrated a nearly uniform magnification of the impedance decrease with pacing (Figure 3).
The main findings of this randomized, controlled trial are: 1) cardiac pacing to accelerate the rate of ventricular contractions significantly reduces contact force variability without altering mean contact force; 2) the achieved reduction in contact force variability with pacing resulted in a higher percentage of time within operator pre-specified contact force goals; and 3) the reduced contact force variability is associated with a 30% greater impedance reduction during radiofrequency ablation.
Durable lesion formation is the hallmark of long-term ablation success. As a result, there is ongoing research to identify means to improve lesion quality. Impedance reduction during ablation has consistently correlated with lesion size in animal models (3,4). However, by its nature, impedance reduction can only be assessed after energy application and, therefore, cannot be used by the operator to decide whether or not to apply energy. In contrast, measurements of contact force offer the operator a tool to predict lesion quality before the application of radiofrequency energy and, thus, offers unique advantages.
Initial studies designed to assess the accuracy of contact force as a tool to predict lesion size have focused predominantly on assessing 1 measure of contact, namely, the mean contact force for the duration of a lesion (10,11). The mean contact force is correlated moderately with impedance reduction (3,9). In addition, studies have demonstrated that, with constant power and prolonged ablation time (>30 s), moderate contact force was moderately proportional to lesion volume (4,7). However, when compared directly, the ability of mean contact force to predict lesion volume was inferior to the predictive accuracy of post-ablation impedance reduction (12). These data suggest that additional measures of catheter–tissue contact beyond mean contact force may better predict lesion formation.
With these data and the advent of clinically available catheters that provide real-time data on contact force, clinical trials were performed to assess if real-time catheter tissue contact data would improve ablation outcomes, in particular for AF. One might expect that, based on the preclinical data outlined herein, mean contact force would be a significant predictor of ablation success. However, this was not the case. In the EFFICAS I trial (Atrial Fibrillation Percutaneous Catheter Ablation with Contact Force Support I), operators were blinded to contact force data during AF ablation in 40 consecutive patients who then underwent repeat assessment at 3 months to determine the frequency and predictors of pulmonary vein reconnection. Analysis of contact force data from the initial ablation procedure revealed that minimum contact force was a significantly better predictor of pulmonary vein isolation when compared with mean contact force (13,14). Similarly, in the SMART-AF Trial (THERMOCOOL SMARTTOUCH Catheter for the Treatment of Symptomatic Paroxysmal Atrial Fibrillation), 160 patients with paroxysmal AF were randomized to pulmonary vein isolation with and without contact force data prospectively available to guide ablation. This trial found no correlation between mean contact force and freedom from AF at 1 year (5). Specifically, when mean contact force values ranged between 7 and 10 g, long-term ablation success was greatest, whereas contact force values that were consistently low (<7 g) or consistently high (>10 g) resulted in higher recurrence rates (15). Similar to EFFICAS I trial, the most significant predictor of AF ablation success was time spent within the pre-ablation operator-specified ideal range of contact force, rather than mean contact force.
With the knowledge that contact force range, rather than mean contact force, better predicts AF ablation success, a multicenter, randomized, controlled trial (1:1 randomization of contact force on vs. off) was pursued with disappointing results (16). Despite ablation prospectively prescribed to be performed with a contact force goal of 5 to 40 g, the use of real-time contact force data failed to improve long-term AF ablation results. However, a deeper examination of these data reveals that, despite this pre-specified goal, contact force was maintained within 5 to 40 g during only 80% of ablation time. These sequential studies, therefore, demonstrate that although maintaining contact force within a given range is ideal to achieve long-term ablation success, operators are unable to achieve this during 20% of ablation time. In light of this finding, identifying mechanisms to improve catheter–tissue contact goals may be critically important to improving ablation outcomes.
The results from our randomized, controlled trial identify a new mechanism to overcome this challenge of maintaining consistent contact force. Pacing to accelerate ventricular rate reduced contact force variability, optimizing catheter–tissue contact. Pacing at an accelerated ventricular rate increased the percentage of time within the pre-specified contact force range. This finding was mirrored by a reduction in the SD and the overall range of the contact force. Of note, this condition was achieved while maintaining an unchanged mean contact force. These finding were nearly uniform across all areas of the left atrium and the cavotricuspid isthmus. Concomitant with this pacing-induced reduction in contact force variability was a significant improvement in impedance reduction during ablation.
One can hypothesize possible explanations for the regional variabilities found in this study. Native conduction while the catheter is located in the mid posterior left atrium likely results in more significant motion with each contraction. This may result in the catheter catching on a ridge-like extension of the carina, thus increasing the maximal contact force. With regard to the increased minimum contact force at the right superior posterior aspect of the left atrium, the right side (when compared with the left side) of the left atrium previously has been shown to be more significantly affected by ventricular contraction (8). The tip of the ablation catheter is frequently perpendicular to the tissue at the right superior posterior aspect of the left atrium. Ventricular contraction results in the downward displacement of the mitral annulus toward the ventricles and the subsequent downward pull of the left atrium. This direction of left atrial displacement is directly at a pointing ablation catheter. As such, this combination of catheter orientation relative to the direction of atrial displacement likely leads to magnified changes in the minimum contact force.
Beyond improving lesion quality with pacing at an accelerated heart rate, pacing also reduced maximal contact force achieved during ablation lesions. Previous studies have demonstrated that elevated contact force values predict steam pops, thrombus formation, and perforation (3,4,17,18). Pacing, therefore, may be a novel tool to lower the risk of these complications during ablation.
Other techniques to improve catheter stability, namely, jet ventilation and the use of steerable sheaths, have been associated with improved AF ablation outcomes (19,20). Clinical corroboration of pacing as a means to improve stability is evident in the use of rapid transvenous pacing for over a decade to stabilize cardiac motion during balloon aortic valvuloplasty, reducing both balloon dislodgement rates and aortic insufficiency (21,22). Recognizing that approximately 29% of contact force variability can be attributed to systolic–diastolic heart motion, interventions aimed at modulating cardiac contraction, like pacing, may have a dramatic effect on catheter tissue contact, lesion formation, and clinical outcome (8).
A significant strength of the study is the use of a randomized process with participants serving as their own controls. However, our study has a number of limitations. Acceleration of cardiac rate to 500 ms was selected based on a clinical assumption that patients would be able to tolerate this degree of heart rate elevation for an extended period of time. It is unknown if faster heart rates would alter catheter–tissue contact in unexpected ways. To perform this study, operators had to be aware of contact force values in real time. As such, it is possible that operators could be biased to try to improve outcomes of one of the arms. However, the data analysis demonstrates that there was no difference in the mean contact force or force–time integral between the 2 arms, strongly suggesting that the operators pursued the study in an unbiased fashion. Although multiple studies have demonstrated that impedance reduction strongly predicts lesion size, there are no histologic data from our study to confirm that lesion formation was improved with the greater decrease in impedance associated with pacing during ablation. Further histologic and clinical outcome studies, including cohort studies or randomized, controlled trials, are needed to corroborate the usefulness of this tool for both acute and long-term ablation outcomes and complications. We excluded patients who presented to the EP laboratory with AF. It remains unclear how catheter stability would be affected by R-R variability during AF when compared with ventricular based pacing in AF. Similarly, our study does not include an assessment of the effect of the location of atrial pacing and the potential effect on the relative timing of atrial and ventricular contraction and its subsequent effects on cardiac and catheter motion. Finally, respiration has been shown to alter contact force independent of cardiac contraction (22). We maintained identical ventilation settings for all patients within the study and chose a mode of ventilation most typically used in electrophysiology laboratories to broaden the potential applicability of this study; however, cardiac contraction and respiration may modify each other to further alter catheter–tissue contact.
Pacing-induced heart rate acceleration reduces catheter–tissue contact variability, increases the probability of achieving pre-specified catheter–tissue contact endpoints, and enhances markers for lesion formation. Modulating the heart rate to improve catheter–tissue contact offers a new approach to optimize ablation outcomes and minimize complications.
COMPETENCY IN MEDICAL KNOWLEDGE: It is known that AF recurrence after ablation is most frequently the result of pulmonary vein reconnection. In addition, catheter instability has been shown to predict AF recurrence. This study demonstrates that pacing-mediated heart rate acceleration improves catheter–tissue contact and stability during left and right atrial ablation.
TRANSLATIONAL OUTLOOK: Pacing-mediated heart rate acceleration improves catheter–tissue contact and thereby may enhance the creation of durable ablation lesions and increase AF ablation success rates.
Dr. Aizer serves as a consultant for Biosense Webster. Dr. Barbhaiya has received speaking fees/honoraria from Medtronic, Inc., Abbott, Inc., and Zoll, Inc. Dr. Chinitz serves as a consultant for Biosense Webster. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
All authors attest they are in compliance with human studies committees and animal welfare regulations of the authors’ institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the JACC: Clinical Electrophysiology author instructions page.
- Abbreviations and Acronyms
- atrial fibrillation
- Received July 24, 2017.
- Revision received December 8, 2017.
- Accepted December 11, 2017.
- 2018 American College of Cardiology Foundation
- Koruth J.,
- Dukkipati S.,
- Gangireddy S.,
- et al.
- Ikeda A.,
- Nakagawa H.,
- Lambert H.,
- et al.
- Natale A.,
- Reddy V.,
- Monir G.,
- et al.
- Sarkozy A.,
- Shah D.,
- Saenen J.,
- et al.
- Weiss C.,
- Antz M.,
- Eick O.,
- Eshagzaiy K.,
- Meinertz T.,
- Willems S.
- Everett T.H.T.,
- Lee K.W.,
- Wilson E.E.,
- Guerra J.M.,
- Varosy P.D.,
- Olgin J.E.
- Rozen G.,
- Ptaszek L.,
- Zilberman I.,
- et al.
- Neuzil P.,
- Reddy V.,
- Kautzner J.,
- et al.
- Reddy V.,
- Pollack S.,
- Lindsay B.,
- et al.
- Ullah W.,
- McLean A.,
- Tayebjee M.,
- et al.
- Perma F.,
- Heist E.K.,
- Danik S.B.,
- Barrett C.D.,
- Ruskin J.N.,
- Monsour M.
- Daehnert I.,
- Rotzch C.,
- Wiener M.,
- Schneider P.
- Ertugul I.,
- Karagoz T.,
- Celiker A.,
- Alehan D.,
- Ozer S.,
- Ozkutlu S.