Author + information
- Received April 8, 2016
- Revision received July 5, 2016
- Accepted July 6, 2016
- Published online November 1, 2016.
- Vivek Y. Reddy, MDa,∗ (, )
- Scott Pollak, MDb,
- Bruce D. Lindsay, MDc,
- H. Thomas McElderry, MDd,
- Andrea Natale, MDe,
- Charan Kantipudi, MDf,
- Moussa Mansour, MDg,
- Daniel P. Melby, MDh,
- Dhanunjaya Lakkireddy, MDi,
- Tzachi Levy, BScj,
- David Izraeli, MSck,
- Chithra Sangli, MSk and
- David Wilber, MDl
- aIcahn School of Medicine at Mount Sinai, New York, New York
- bFlorida Hospital, Cardiovascular Research, Orlando, Florida
- cCleveland Clinic Foundation, Cleveland, Ohio
- dUniversity of Alabama, Birmingham, Alabama
- eTexas Cardiac Arrhythmia Research Foundation, Austin, Texas
- fPiedmont Heart Institute, Atlanta, Georgia
- gMassachusetts General Hospital, Cardiac Arrhythmia Center, Boston, Massachusetts
- hAbbott Northwestern Hospital, Minneapolis, Minnesota
- iUniversity of Kansas, Mid-America Cardiology, Kansas City, Kansas
- jBiosense Webster (Israel) Ltd., Yokneam, Israel
- kBiosense Webster Inc., Diamond Bar, California
- lLoyola University Medical Center, Maywood, Illinois
- ↵∗Reprint requests and correspondence:
Dr. Vivek Y. Reddy, Helmsley Trust Electrophysiology Center, Icahn School of Medicine at Mount Sinai, One Gustave L. Levy Place, Box 1030, New York, New York 10029.
Objectives This study sought to assess the correlation between catheter and tissue contact force (CF) stability and 12-month clinical success for atrial fibrillation (AF) ablation.
Background The SMART-AF (Thermocool Smarttouch Catheter for the Treatment of Symptomatic Paroxysmal Atrial Fibrillation) multicenter trial provided a robust dataset of AF ablation procedures, using the CF sensing ablation catheter.
Methods CF and CF stability were correlated with 12-month success for drug-refractory symptomatic AF ablation. CF stability was assessed by stability of ablation parameters (CF, time, location stability) over 3-dimensional electroanatomic maps of pulmonary veins (PVs) using a new proprietary software module and the percentage of time within investigator-selected CF ranges. Available data for potential “PV gaps” were retrospectively identified when stability criteria were not met and were correlated with 12-month success.
Results Average CF categories of 0 to 10, 10 to 20, and >20 g were associated with 12-month success rates of 90%, 70%, and 70%, respectively; thus, higher average CF did not correlate with treatment success. An exploratory univariate analysis showed significantly higher success rates with a CF of 6.5 to 10.3 g than with <6.5 g (odds ratio: 2.95; 95% confidence interval: 1.13 to 7.72; p = 0.028) but a CF >10 g did not improve success. When stable CF was applied ≥73% of the time within the preselected CF range, success improved. A receiver operating characteristic curve analysis revealed that PV gaps exceeding 10.6-mm distance significantly correlated with 12-month failure.
Conclusions In the SMART-AF trial, CF stability with sufficient CF was most predictive of optimal 12-month success. (Thermocool Smarttouch Catheter for the Treatment of Symptomatic Paroxysmal Atrial Fibrillation [SMART-AF]; NCT01385202)
- catheter stability parameters
- contact force
- paroxysmal atrial fibrillation
- pulmonary vein isolation
- radiofrequency catheter ablation
Radiofrequency (RF) catheter ablation is an established treatment for drug-refractory paroxysmal atrial fibrillation (1–3). Pulmonary vein isolation (PVI) is a widely used acute procedural endpoint; however, clinical recurrences are common (2). This is partly due to the relatively poor indicators that are currently available to predict ablation adequacy: electrogram diminution and impedance drop (4–9). The correlation between electrode and tissue contact and lesion generation has been evaluated in vivo and in vitro studies (5–7). Clinical studies using contact force (CF)-sensing catheters have reported various degrees of correlation between CF and clinical outcome (4,10–12). There is broad agreement that CF needs to be carefully monitored because insufficient CF is ineffective but excessive CF may result in complications such as esophageal injury, “steam pop,” thrombus formation, and cardiac tamponade due to atrial perforation (13–16).
The prospective, multicenter, nonrandomized, single-arm SMART-AF (Thermocool Smarttouch Catheter for the Treatment of Symptomatic Paroxysmal Atrial Fibrillation) trial evaluated the safety and effectiveness of the CF-sensing open-irrigated RF Thermocool Smarttouch catheter (Biosense Webster Inc., Diamond Bar, California) for RF ablation of drug-refractory symptomatic paroxysmal atrial fibrillation (4). The probability of freedom from arrhythmia recurrence was greater when the investigators stayed within their preselected target CF ranges for a greater proportion of time during AF ablation (4). Although the importance of catheter-tissue CF during AF ablation is recognized (4,10,11), the influence of catheter stability on ablation outcome has never been determined.
This analysis of data from the SMART-AF trial reevaluated the correlation between specific levels and categories of CF and 12-month clinical success. CF and location stability of CF were examined as additional predictors of clinical success. Furthermore, stability criteria, including location stability and minimum time and force parameters, were used to determine PV gaps in ablation and their correlation with clinical success.
Study design and patients
The present study is an analysis of the CF stability data from the SMART-AF trial (NCT01385202), details of which have been previously published (4). Briefly, patients with at least 3 symptomatic AF episodes within 6 months and at least 1 documented AF episode within 1 year of enrollment and unresponsive to at least 1 antiarrhythmic drug (class I or III or atrioventricular nodal blocker) were evaluated at 1, 3, 6, 9, and 12 months post-ablation, with a 3-month blanking period for anatomic and electrical remodeling of the left atrium. Electrocardiograms were recorded at all follow-up visits, with transtelephonic monitoring during the 9-month post-blanking period. The study protocol was approved by the respective ethics committees or institutional review boards, and all patients provided written informed consent.
Catheter ablation and CF measurements
The 3-dimensional electroanatomic mapping of the left atrium and PVs was performed using the Carto 3 system (Biosense Webster, Inc.). A circumferential set of lesions was created for PVI, which was confirmed by entrance block.
A CF working range was not prescribed but was instead preselected by each investigator based on experience. CF applied at different PV segments (Figure 1) was analyzed. CF was averaged from 20 measurements over 1 s. A 10-s running graph presented real-time data. CF measurements were sampled at 50-ms intervals (20 Hz) during RF application, translating into 90,000 or more data points per case. Each data point was analyzed to determine whether it was within the prespecified working range for each case. The percentage of time the investigator was within the CF working range was calculated by the number of data points within the [working range/total data points × 100].
Carto Visitag module
Complete CF and ablation stability data for a subset of SMART-AF patients from 10 centers, including roll-in patients (the initial cases performed to allow operators to get accustomed to the new catheter without inclusion in the primary analysis), with available data (n = 40) were further analyzed retrospectively using the Carto Visitag module (Biosense Webster, Inc.). The module displays ablation parameters such as power, impedance, stability, and time at each specific ablation location. The combination of the module with the Carto 3 system permits alteration of ablation parameters in real time to maintain catheter stability within a user-selected range. Retrospectively, the parameters (i.e., stability criteria) used for the present analysis were location stability of 2 mm, giving consideration to respiratory movement and heartbeat; application of RF energy for a minimum of 10 s at each ablation site; and a CF of 5 g, which is considered a minimum effective touch. “PV gaps” were locations where stability criteria were not met.
Previous studies have found a positive association between increased CF and freedom from AF recurrence at 12 months (10). Therefore, the association between CF value and 12-month success was evaluated. Fisher’s exact test was used to test for the association of freedom from AF with the average CF stratified as 0 to 10, 10 to 20, and >20 g. Effectiveness was also descriptively and graphically evaluated for the median, 25th, and 75th percentiles and minimum and maximum CFs. Due to the skew in the distribution of CF data, the 25th percentile of the CF for each patient was used to evaluate the association between CF and effectiveness. Stratum membership based on observed tertiles from the distribution of the 25th percentile of the CF was entered into a logistic regression model as the predictor variable. Unlike the previous stratification in which some strata were sparse, the stratification based on observed tertiles allowed for roughly equal representation of subjects in each CF category, providing better resolution for evaluation of the association between CF and clinical success.
In regard to CF stability, analyses were performed to evaluate effectiveness for various cut points on the percentage of time investigators stayed within their preselected CF range. In addition, analysis of the Carto Visitag module was performed in a subset of 40 patients in the effectiveness cohort of the SMART-AF trial (4), for whom CF backup files were available. In addition, receiver operating characteristic (ROC) curve analysis of sensitivity versus specificity determined the cutoff for the width between the centroids of adjacent PV gaps and the correlation between the cutoff for PV gaps, and 12-month outcome was assessed.
Demographics and baseline characteristics
The Carto Visitag module subanalysis included a subset of 40 patients with available complete CF backup data files. This Carto Visitag module subset (Table 1) was similar at baseline to the full SMART-AF effectiveness cohort, as previously reported (4). The 12-month clinical success of freedom from AF of this 40-patient cohort with available data was not significantly different from that of the remaining patients in SMART-AF (log-rank test p = 0.108).
Figure 2A shows the average CF applied at the different PV sites. The highest CF was recorded at the right anterior wall and right inferior segment for the right PV and at the left roof for the left PV. The lowest CF values were recorded at the left carina, followed by the right carina. Overall, the CF was higher for the right than for the left PVs. An example of the color-coded force map superimposed on the image from the Carto 3 system is presented in Figure 2B.
Correlation of CF with success at 12 months
Clinical success was not associated with average CF when categorized as 0 to 10, 10 to 20, and >20 g (p = 0.134) (Figure 3). Specifically, the use of a higher average CF did not translate into better 12-month treatment success. When analyzed by using the 25th percentile of the CF, the odds of success were significantly higher in the second tertile (6.5 to 10.3 g) than in the first tertile (CF: <6.5 g; odds ratio: 2.95; 95% confidence interval: 1.13 to 7.72; p = 0.028). However, there were no significant differences between the odds of success of the third (CF: >10.3 g) and those of the first tertiles (CF: <6.5 g) (Table 2).
The investigator-selected CF working range varied from a low of 4 g to a high of 60 g (Table 3). A CF working range of 5 to 40 g was selected in two-thirds of the procedures for which CF data were available. There was an increase in success rate observed with an increase in the proportion of time investigators spent in their selected CF ranges (Figure 4). Specifically, success rates increased when the applied CF was within the investigator-selected CF range more than 73% of time.
Correlation between Carto Visitag module-defined CF stability and PV gaps with success at 12 months
Among the 40 patients with CF and ablation stability data for analysis, clinical success at 12 months was noted in 58% of patients (23 of 40). On the basis of predefined stability criteria and ROC curve analysis, a PV gap was defined as a distance of >10.6 mm between the centers of adjacent ablation sites (Figure 5A). Success at 12 months was significantly higher in patients without PV gaps than in those with PV gaps (93.8% vs. 33.3%, respectively) (Figure 5B).
Long-term AF recurrences post-ablation are common, and early recurrences appear to be a strong predictor of late recurrences (17). The durability of PVI is critical to freedom from AF recurrence post-ablation. Pre-clinical studies showed that the CF between the tip of the ablation catheter and the target tissue determined the lesion characteristics and therefore the effectiveness of the procedure (13,14). However, clinical studies evaluating the influence of CF on treatment effectiveness up to 1 year have yielded mixed results. In the TOCCATA (Touch+ for Catheter Ablation) study, which used a different CF-sensing catheter technology, the average CF used for ablation correlated with clinical outcomes; an average CF >20 g provided the best arrhythmia control, whereas average CF of <10 g was associated with clinical failure (10). In contrast, the SMART-AF trial found no correlation between higher average CF and treatment effectiveness (4).
The current analysis of CF data from the SMART-AF trial did not find an association between 12-month clinical success and average CF when categorized as 0 to 10, 10 to 20, and >20 g, which is consistent with a recent preclinical finding that there were no differences in lesion size, quality, or transmurality with ablation, using high (>20 g) or low (<10 g) CF (7). However, further analyses of the data suggested a nonlinear relationship of effectiveness with the CF; optimal success (82.6%) was observed with a CF range of ∼6 to 10 g. A further increase in the CF to >10 g was not associated with additional improvement in the odds of effectiveness. The differences between our analysis findings and those of the previous study using a different CF-sensing catheter (10) may be attributed to operator experience differences between the 2 studies, potential differences in the physical properties of the 2 CF-sensing catheters (e.g., catheter stiffness), or user instructions and different mapping systems used. Another important, and perhaps likely, explanation is the state of our understanding of the effect of force on lesion biophysics at the time of commencement of each clinical trial, that is, because TOCCATA (10) was performed at a time when the importance of maintaining a minimum CF was not appreciated. However, by the time of the SMART-AF trial (4), not only was the importance of maintaining a minimum force accepted, but it was also appreciated that instances of low tip-tissue contact force can be compensated by increasing the amount of delivered power. Given these evolving conditions, it becomes less surprising that there were no changes in clinical outcome as a function of the overall average contact force.
The SMART-AF trial previously showed more than 4-fold increased likelihood of clinical success when investigators stayed within their pre-selected CF working ranges ≥80% of the time (4). Although higher average CF (>10 g) during RF ablation did not correlate with 12-month success, an increase in the time within investigator-selected CF ranges during most of the procedure (for more than ∼73%) did. Taken together, the data suggested that stable catheter-tissue contact with sufficient minimum CF to enable contact, rather than achieving higher than necessary CF, contributes to optimal clinical success.
Variations in CF application between different anatomic PV segments in SMART-AF were observed as in the TOCCATA study. The latter also showed that intermittent ablations coinciding with diastole were more frequent with low CF (10). A lack of stable contact between the catheter tip and target tissue may produce suboptimal lesions or gaps that lead to eventual PV reconnection (18). High CF, on the other hand, is associated with increased frequency of complications such as damage to adjacent anatomic structures, including cardiac perforation (13–16,19). In the SMART-AF trial (4), there was a nonsignificant trend toward higher incidence of tamponade with high CF during RF application. The correlation was of borderline significance (p = 0.0581), likely due to the small number of events. A real-time display of various ablation parameters during the procedure would allow the operators to tailor the CF stability setting to create durable lesions while minimizing the occurrence of complications.
The Carto Visitag module became available after the SMART-AF trial (4) was completed. Retrospective analysis using predefined stability criteria within the module identified the PV gap (defined as the distance between the centroids of adjacent lesions, as identified by the Carto Visitag software) as a potential predictor of clinical success. These data reinforced the finding that stable CF applied along continuous ablation lines result in greater clinical success.
Factors such as the topography of the lesion path, PV gaps, CF and its stability, and duration of RF energy application determine the durability of ablation lesions. It was previously shown that PV reconnection rates are significantly reduced by providing the operators with real-time CF information during the ablation procedure (20). This study demonstrated that location stability of CF is a critical factor in addition to the absolute magnitude of CF.
The analysis using Carto Visitag module stability criteria was retrospective in nature because this module was not available at the time of the SMART-AF trial (4) and should be considered hypothesis generating only. In addition, the number of cases with backup files for this analysis was limited. Analyses identifying the nonlinear association of effectiveness with CF were exploratory and post hoc. Furthermore, the lack of integrated computation taking into consideration application time and power data makes predictions for true full/complete lesions difficult, whereas the limited numbers preclude definitive conclusions about the gap size for vein isolation. These results will need to be confirmed in further study where ablation is prospectively guided by the Carto Visitag module with a minimum CF setting.
Our results suggest there is a nonlinear relationship between CF and optimal clinical outcome and that CF stability with sufficient CF to enable contact, rather than achieving higher average CF, is the key to creating durable ablation lesions and optimizing long-term success.
COMPETENCY IN MEDICAL KNOWLEDGE: In view of frequent AF recurrences due to PV reconnection following ablation procedures, real-time assessment and adjustment of CF parameters may help operators create durable lesions, thereby improving clinical outcomes. The subanalysis of SMART-AF reveals that the most important determinant of clinical outcome is not the overall average contact force but rather catheter stability and lesion contiguity.
TRANSLATIONAL OUTLOOK: Integration of electroanatomic mapping with CF parameters by using the Carto Visitag module may improve the stability of CF application, create durable lesions, and decrease the need for repeat procedures.
The authors thank the SMART-AF trial investigators (Online Table 1) and the following individuals who contributed to the study and to statistical analysis and provided editorial assistance: Robert Stagg, Doron Ludwin, Garth Constantine, Lee Ming Boo, and Yulin Zhang. Additional editorial support was provided by Annirudha Chillar, Cactus Communications. The authors retained full control of manuscript content development, including revisions. The authors had full access to sponsor-provided data that were collected during execution of the study protocol.
For a supplemental table, please see the online version of this article.
The study was sponsored by Biosense Webster, Inc. Dr. Lindsay is a compensated member of Biosense Webster scientific advisory board. Dr. McElderry is a consultant for and has received research support from Biosense Webster and Boston Scientific; and is a consultant for St. Jude Medical and Medtronic. Dr. Natale is a compensated member of the advisory boards for Janssen, Medtronic, Boston Scientific, St. Jude Medical, and Biosense Webster. Dr. Mansour is a consultant for Biosense Webster, St. Jude Medical, Biotronik, and Medtronic; and has received research grants from St. Jude Medical and Biosense Webster. Dr. Melby is a compensated speaker and consultant for Biosense Webster. Dr. Levy is an employee of Biosense Webster. Dr. Izraeli is an employee of Biosense Webster. Dr. Sangli is an employee of and holds stock in Biosense Webster. Dr. Wilber is a consultant for Biosense Webster, Medtronic and Abbott, and the American College of Cardiology Foundation. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
Jon Kalman, MBBS, PhD, served as Guest Editor for this paper.
- Abbreviations and Acronyms
- atrial fibrillation
- contact force
- LIPV anterior wall
- left carina
- left inferior
- left inferior pulmonary vein
- left posterior wall
- left pulmonary vein
- left roof
- LSPV ridge
- left superior pulmonary vein
- left ventricular ejection fraction
- pulmonary vein
- pulmonary vein isolation
- right anterior wall
- right carina
- right inferior
- right inferior pulmonary vein
- right posterior wall
- right pulmonary vein
- right roof
- right superior pulmonary vein
- Received April 8, 2016.
- Revision received July 5, 2016.
- Accepted July 6, 2016.
- American College of Cardiology Foundation
- Stabile G.,
- Bertaglia E.,
- Senatore G.,
- et al.
- Calkins H.,
- Kuck K.H.,
- Cappato R.,
- et al.
- Jaïs P.,
- Cauchemez B.,
- Macle L.,
- et al.
- Natale A.,
- Reddy V.Y.,
- Monir G.,
- et al.
- Williams S.E.,
- Harrison J.,
- Chubb H.,
- et al.
- Ouyang F.,
- Antz M.,
- Ernst S.,
- et al.
- Ouyang F.,
- Tilz R.,
- Chun J.,
- et al.
- Kautzner J.,
- Neuzil P.,
- Lambert H.,
- et al.
- Yokoyama K.,
- Nakagawa H.,
- Shah D.C.,
- et al.
- Cappato R.,
- Calkins H.,
- Chen S.A.,
- et al.
- Nakagawa H.,
- Kautzner J.,
- Natale A.,
- et al.