Determinants of Atrial Lesion Maturation during Radio Frequency Ablation Using Localized Tissue Electrograms

DOI: 10.19102/icrm.2014.050402

¹BOAZ AVITALL, MD, PhD, FACC, FHRS, ¹PIOTR HORBAL, BS, ¹DAVID VANCE, MD and ²JOSEF KOBLISH, BS

¹University of Illinois Chicago, Chicago, IL, USA

²Boston Scientific Corporation, Natick, MA, USA

Download PDF

Follow Us >> Tweet

ABSTRACT. Introduction: We hypothesize that titration of radiofrequency (RF) ablation time to the lowest level of mini electrode (ME) electrogram (EGM) amplitude will create transmural atrial lesions and deep ventricular lesions. Methods: A novel 8-mm non-irrigated catheter was equipped with a 3×0.8 mm ME radially distributed and embedded 2 mm from the catheter tip, in addition to three standard ring electrodes. A total of 51 atrial and 22 ventricular lesions were placed in four normal sinus rhythm canines (30–35 kg) using temperature control RF target of 65°C and maximum power of 65 watts. RF ablation was stopped at the lowest level of ME EGM amplitude. EGM and pacing threshold (TH) measurements were recorded pre- and post ablation at the 8-mm ring electrode and between the ME. Results: The ME local EGM amplitudes and pacing TH were significantly different than those of the 8-mm ring electrode, as were the respective responses of each to RF ablation. Average lesion depth in the atrium and ventricle was 2±0.7 mm and 5±2.1 mm with average RF time of 25.2±8.2 s and 32.1±17.9 s, respectively. This resulted in 96% of the atrial lesions being transmural. No impedance rise, steam pops, or char formation were documented and pathological assessment revealed no unintended injuries. Conclusions: This study demonstrated that the use of localized EGM signals from the ME and titration of RF time to the lowest electrogram amplitude recorded from the ME resulted in transmural atrial lesions and deep ventricular lesions, while maintaining efficacy and safety.

KEYWORDS. atrial fibrillation, lesion monitoring, mini electrodes, radiofrequency.

Financial Support for this research was provided by Boston Scientific Corporation (BSC), Natick, MA. Dr. Avitall reports he is a consultant to BSC and Mr. Koblish is an employee of BSC.
Manuscript received February 10, 2014, Final version accepted March 4, 2014.

Address correspondence to: Dr. Boaz Avitall, University of Illinois Chicago, 840 South Wood St. Suite 922, Chicago, IL 60612, 312-996-9090.
E-mail: bavitall@uic.edu

Introduction

Unlike ventricular tissue, the atria consist of thin tissue with little exposure to intramyocardial blood flow or cooling from epicardial vasculature. As a result, the likelihood of extracardiac damage increases with prolonged radiofrequency (RF) application. An accurate assessment of transmural lesion maturation would reduce the risk of extracardiac injury while still maintaining ablation efficacy.

Several publications have shown a local electrogram (EGM) amplitude reduction greater than –50% during RF application is associated with the formation of transmural atrial lesions;^1–4 however, the monitored bipolar EGM with the 8-mm tip to the first recording ring extends beyond ablated tissues, reducing the effect of RF application on the monitored EGM. We hypothesize that 8-mm RF ablation catheters equipped with additional mini electrodes (MEs) may be used to titrate RF ablation based on the maximal reduction of EGM recorded from the ME, thereby maintaining the efficacy and safety of RF ablation. This study was designed to define the lesion dimensions using this approach in both the atria and ventricle.

Methods

Animal care

All procedures were performed in compliance with the American Heart Association's guidelines for animal research and were approved by the Animal Care Committee of the University of Illinois at Chicago. Anesthesia induction with intravenous propofol (0.5 mg/kg) and maintenance with inhaled isoflurane 1.00–1.50% via endotracheal intubation was used for all invasive canine interventions. Invasive hemodynamic monitoring and temperature monitoring was performed throughout the entire experimental protocol and maintained at 38°C using a heating blanket.

Experimental and ablation protocol

Each canine underwent endotracheal intubation. A duodecapolar catheter was inserted into the external jugular vein and positioned in the high right atrium (RA) extending into the coronary sinus (CS). Chamber anatomy and catheter positioning were confirmed using the EnSite NaVx 3D (St. Jude Medical, St. Paul, MN) navigation system. Using fluoroscopic and transesophageal echocardiographic guidance, transseptal puncture was performed. The lesions were guided by 3D navigation posterior-anterior (PA), left anterior oblique (LAO), right anterior oblique (RAO) fluoroscopic imaging. In addition, lesions were placed in clearly defined anatomical landmarks: superior vena cava (SVC), inferior vena cava (IVC), RA to right ventricle (RV) isthmus junction, right atrial appendage (RAA), the distal sleeve and base of each pulmonary vein (PV), posterolateral left atrium (LA) to left ventricle (LV) junction, and left atrial appendage (LAA). Following completion of the atrial lesions, LV and RV lesions were then administered sequentially guided by the 3D navigation and fluoroscope to the posterolateral wall, mid-posterolateral wall, apex, mid-septum, and high septum in both ventricles. Individual lesions were placed at least 25 mm apart in each of the cardiac chambers. These lesions were guided by a combination of 3D navigation and PA, LAO, and RAO fluoroscopic imaging.

Lesions were placed with an 8-mm non-irrigated catheter equipped with three 0.8-mm MEs radially distributed at 1.2-mm intervals and embedded 2 mm from the catheter tip in addition to the three standard ring electrodes (Figure 1). EGM and pacing threshold (TH) (1 ms pulse width and maximum amplitude of 20 mA) measurements were recorded pre- and post ablation between the 8-mm tip to the first distal ring electrode and between the three MEs in an overlapping fashion (1-2, 2-3, 3-1). RF generator settings for the 8-mm non-irrigated ME catheter were set to a temperature mode target of 65°C and maximum power of 65 W. EGMs were recorded pre, during, and post ablation using the EP-Med Workmate System (St. Jude Medical, St. Paul, MN) with the filter setting at 0.5–500 Hz. RF power was applied to maximal EGM reduction and 5 s of plateau.

The data from each pair of MEs was tabulated and analyzed by the extent of EGM change from pre- to post ablation and was defined as level 1 (maximal decrease in EGM amplitude) to 3 (least decrease in amplitude). This arrangement of the data was carried out to account for the varying degrees of ME contact with the tissue as a result of the 8-mm angle of contact with the cardiac tissues as the catheter is moved.

A total of 51 atrial lesions and 22 ventricular lesions were placed in four normal sinus rhythm canines (30–35 kg). Lesion times were titrated to the maximal loss of ME EGM amplitude.

At the conclusion of the study the heart, lung, and esophagus were excised and examined for injury. All cardiac lesions were examined for transmurality by assessing the epicardial surface for clearly demarcated lesions. Each lesion was marked, photographed, and matched to both the ablation 3D navigation positioning and the ablation sequence. The hearts were immersed in tetrazolium blue, and lesions were then visually re-examined, re-photographed, and measured for length and width. The center of each lesion was sectioned and the depth of ablated tissue was measured using digital calipers.

Single factor analysis of variance (ANOVA) was applied to determine if there were significant differences between mean values obtained from each of the electrode pair recordings. All data were expressed as mean values±1 standard deviation. The null hypothesis was rejected at the level p<0.05.

Results

Atrial tissues

Pre-ablation the average pacing TH at ME level 1 was significantly lower than the 8-mm ring electrodes (1.4±2 versus 2.8±1.6 mA, p<0.05). Post ablation, ME level 1 TH rise was markedly greater than in the 8-mm ring and significantly greater than ME levels 2 and 3. Higher baseline TH and lower levels of TH rise were noted in levels 2 and 3 (Figure 2).These changes correspond to the EGM recordings pre- and post ablation as shown in Figure 3.

Pre-ablation, the EGM amplitude recorded from the 8-mm ring was significantly greater than that recorded by the ME. Among the MEs, level 3 was significantly lower than levels 1 and 2. Post RF application, all electrode configurations exhibited amplitude reduction; however, the reduction recorded by the ME was significantly greater than the 8-mm ring. The percent reduction (Figure 2B) pre- to post ablation recorded from the 8-mm ring was –37.6±36%, whereas the reduction recorded with the ME level 1 was –85.8±10% (p<0.001). The extent of the amplitude decrease varied significantly among the three pairs of MEs. The percent reduction in amplitude for ME level 2 was –77.4±16% (p<0.05 versus level 1) and for ME level 3 was –65.3±23% (p<0.001 versus level 1). Average lesion depth in the atrium was 2±0.7 mm with average RF time of 25.2±8.2 s. This resulted in 96% of the atrial lesions being transmural.

An example of the tissue pathology and the corresponding EGM changes is shown in Figure 4. The lesion placed in the RAA across trabeculated myocardium was generated with 11 watts of power applied for 28 s achieving 65°C. The Tetrazolium blue-stained tissues (Figure 4A) show that the lesion is clearly demarcated on both the epicardial and endocardial tissues. The figure also illustrates the marked reduction of EGM amplitude recorded by the ME in response to the destruction of tissues in contact with the 8-mm ablation electrode (Figure 4B). Whereas, the 8-mm ring recording exhibited –15% amplitude reduction, the ME EGM reduction was –91.4, –90.9, and –86.1%. It is important to note that ventricular recordings are also present on both the 8-mm ring and the ME due to the RAA overriding the RV outflow epicardial tissues; however, following titration of the RF application time to the maximal reduction of the ME EGM (28 s), neither ventricular EGM reduction nor ventricular epicardial injury were noted.

A series of five consecutive lesions along the intracaval line are shown in Figure 5. All the lesions were clearly identified on both endocardial (Figure 5A) and epicardial (Figure 5B) surfaces of the atria. Table 1 provides the TH and EGM amplitude pre- and post ablation, RF power, 8-mm tip recorded peak temperature during ablation, and the RF application time in seconds for each of the lesions. Lesion 1 was placed at the junction between the SVC and the RA sleeve. The pacing TH were higher and the recorded EGM were lower at this location versus the 4 other lesions and the EGM percent reduction varies dramatically: –81.2%, –69.2, and –34.9%. Lesion 1 also required high RF power and achieved a lower temperature due to exposure to high blood flow and the ablation tip straddling the SVC and RA myocardial sleeve. EGM recorded at each of the right atrial anatomical locations is shown in Figure 5C. Whereas lesions 2 and 3 were placed on smooth non-trabeculated myocardium, lesions 4 and 5 were positioned between the trabeculi adjacent to the crista terminalis. The 8-mm tip to ring electrode changes in TH could not be relied upon to define the formation of a lesion; whereas the ME rise in TH was dramatic and maximal at the pair that are in the best contact with the myocardium. The EGM reduction recorded from the ME post ablation was far greater than those recorded from the 8-mm ring. The marked increase in the ME TH and the parallel reduction in the ME EGM are associated with transmural atrial lesions such as those shown in Figure 5A,B.

Ventricular lesions

Similar to the pacing TH recorded in the atria, the ME pacing THs are significantly lower than that of the 8-mm ring. As shown in Figure 6 the average TH recorded from the 8-mm ring is 1.9±1.2 mA, whereas the TH recorded at ME level 1 is 0.8±1.7 mA (p<0.05). Post ablation, ME level 1 pacing TH increased to 8.8±6.7 mA (p<0.001), in contrast the 8-mm ring TH increased from 1.9±1.2 mA to 2.6±1.6 mA (p-NS). The pre-ablation TH for ME level 3 is significantly higher than level 1 and with a lower increase post ablation.

The pre/post ablation EGM changes are shown in Figure 7. Whereas the 8-mm ring electrode exhibited no significant reduction pre- versus post ablation (–10.7±35%), ME level 1 EGM reduction is –80±10%, and level 3 is lower than level 1 (–60±20.6, p<0.001 versus level 1).

In the trabeculated thick ventricular tissues the recorded EGM from the ME can reach 20 mV. Despite the deep lesions created and the reduction of the ME EGM, the post-ablation ME recordings were substantially larger than those recorded in the atria. These observations further substantiate the observation that the ME recordings represent not only superficial electrical activity but also deep tissues with maximal EGM reduction of –92.7% and –85.8% (Figure 8; lesions 32, 33). Lesion 34 is a shallower lesion, measuring 4.7 mm with the maximal EGM reduction of –71.4%.

Atrial and ventricular lesions

EGM, lesion size, and the RF application parameters recorded in the atria and the ventricles are shown in Table 2. The table provides the mean, data range, and the median values for the recorded EGM. Post-ablation measurements from 8 mm to ring recordings were highly variable despite the documentation of a transmural lesion. In contrast, the post-ablation ME data are much less variable, as shown by the standard deviation range and median data. The pre-ablation ventricular versus atrial EGM amplitude is significantly larger at every electrode reflecting the thicker tissues and the likelihood that the catheter is embedded within thick trabeculated myocardium.

The width of the atrial and ventricular lesions were similar (6.5±1.5 versus 6.9±1.6 mm, p-NS), but both the lesion length, and most importantly the depth (second to tissue thickness), were significantly longer and much deeper in the ventricle (atrial length 7.8±2.4 versus ventricular length 9.3±2 mm, and atrial depth 2±0.7 versus ventricular depth 5±2.1 mm). To reach the maximal reduction in EGM amplitude, the RF was applied for a longer time than the atria (atrial ablation time 25.2±8.2 versus ventricular ablation time 32.3±8 seconds, p<0.001) while achieving similar tip temperature.

Safety considerations

Using the EGM reduction as a marker for the termination of RF power, we did not observe steam pops, tissue shredding, atrial perforation, char formation, or any indication of damage to extracardiac tissues such as epicardial ventricular lesions, esophageal lesions, or lung injuries.

Discussion

This study was designed to define the lesion boundaries of ME EGM reduction in both the atrium and the ventricle. Ninety six percent of left and right atrial lesions were transmural, including those in trabeculated myocardium. These findings support the hypothesis that electrical activity recorded from the MEs represents not only superficial endocardial tissues but also intramural tissues as deep as 9.3 mm away from the electrodes. Furthermore, it was documented that transmural atrial ablation can be reliably achieved by titration of the RF application time to the maximal reduction of the focal EGM recorded from the ME. In contrast, monitoring the 8-mm ring recordings is highly unreliable. Examination of the cardiac and extracardiac tissues for any unintended injuries resulted in none being found.

Presently, the 4-mm non-irrigated and irrigated ablation catheters as well as the 8-mm non-irrigated catheters are commonly used for ablation procedures. The use of these catheters significantly limits the ability of the operator to define the endpoints of lesion formation and maturation. Temperature monitoring from the catheter's ablation tip remains a poor indicator of tissue heating because the catheter's tip is cooled by saline irrigation and the 8-mm ablation catheter is cooled via the proximal portion of the ablation electrode that is exposed to the circulation.^5,6 In multiple past publications a reduction in local electrical activity has been proposed to be an indicator of lesion formation.^1–3 The inherent limitation of recording and monitoring the electrical activity from these catheters is that the recording spans a distance of 11.5 mm from the 8-mm catheter tip to the proximal edge of the first ring, and 8 mm from the 4-mm catheter tip to proximal edge of the first ring. This span of recording extends beyond the ablation lesion and introduces greater susceptibility to far field signals. An additional limitation is that the recording is done using uneven recording surface areas reducing the noise rejection of a bipolar recording. The addition of the ME at the circumference of the catheter tip focuses the recording to tissues in contact with the ablation catheter, and uniquely enables the recording of tissues of interest prior and during ablation. A similar concept was proposed long ago but has not been implemented in clinical cardiac electrophysiology.^7–9

Pacing threshold

In this study the pacing TH obtained from the ME pairs was significantly less than that of the 8-mm ring pre- and post ablation. In fact, despite the formation of wide and long transmural lesions in the atria and deep large lesions in the ventricles; the increase in pacing TH and EGM diminution were modest at best on the 8-mm ring recordings, whereas the ME exhibited marked TH and EGM changes. The smaller surface area of the ME compared to an 8-mm electrode yields higher current density depolarizing a smaller tissue mass, leading to a lower pacing TH. This property is likely to improve clinical diagnostic pacing maneuvers prior to ablation.

Determinants of lesion size and depth in the atria and ventricles

In this study, lesions placed in the atria were transmural with an average tissue depth of 2±0.7 mm ranging from 0.8 to 3.6 mm. In the atria the lesion depth was limited by the tissue thickness. The ME with the largest percent amplitude reduction (level 1) measured –85.8±10.8% ranging from –54 to –100%. In contrast, the 8-mm ring percent amplitude reduction was –37.7±36.2, ranging from –2 to –88.7%. These data suggest that ME EGM monitoring is much more reliable to determine lesion formation and maturation than 8-mm ring monitoring. The termination of RF power at the maximal reduction of EGM was associated with transmural atrial lesions. The application of ventricular lesions resulted in an average depth of 5±2.1 mm ranging from 2.1 mm in the RV apex to 9.3 mm in the intraventricular septum. The ME with the largest percent amplitude reduction (level 1) was –80.8±10% ranging from –68.1 to –90.6%. In contrast, the 8-mm ring percent amplitude reduction was –10.7±35.3 ranging from –0.38 to –67.5%. Based on these results one may conclude that ME monitoring for maximal EGM reduction during ventricular ablation represents deep lesion formation and it is also markedly superior to monitoring the 8-mm ring.

Safety considerations

It is not an uncommon occurrence that overextending RF application results in steam pops, tissue shredding, and at times pericardial effusions.¹⁰ Whereas the human LV wall thickness varies from 10 to 18 mm,¹¹ human left atrial thickness ranges between 1.2 and 6 mm, with a mean of 3.7 mm.¹² Consequently, atrial resistance to forces applied by 7–8F ablation catheters results in pouching and compression of the tissue toward extracardiac structures. With increasing ablation power, excessive heating of extracardiac tissues may lead to severe and at times deadly complications, which are the consequence of the poor correlation between the ablation electrode temperature and the tissue temperature.^5,12–15 As such, avoiding extracardiac injury caused by atrial RF lesions and the success of ablation depend on accurate information regarding lesion formation. Since the ultimate goal of ablation is the creation of electrically silent tissues, we have focused on providing the operator with higher resolution electrical activity of the tissue being ablated. This study affirms the use of monitoring the time course of EGM diminution using the ME to define RF lesion efficacy and maturation. The RF application time was titrated to the maximal reduction of the ME EGM and was proven to be effective for the creation of atrial transmural lesions and deep ventricular lesions.

Safety is increased by limiting RF application time to the formation of transmural atrial lesion, thereby decreasing ablation time and reducing the possibility of the lesion to extend to extracardiac tissues. Given the limited thickness of tissues in the atria and the compression of tissues by the ablation catheter, the majority of the recordings from the 8 mm to ring electrodes is far field from the surrounding tissues and not related to the tissue thickness. As a result the 8 mm to ring electrodes maximal EGM reduction is much less sensitive to the ablation. As this investigation has proven, the ME recordings are focused on the ablated tissues and are markedly more sensitive to lesion formation versus the 8 mm to ring electrodes. Furthermore, the EGM recordings and post RF changes reflect intramural tissues as deep as 9.3 mm.

In previous investigations using this catheter technology, the reduction of the EGM post ablation was accompanied by the elimination of the high-frequency component of the local cardiac depolarization and the emergence of lower frequency spectra likely associated with current of injury that replaces the fast high-frequency depolarization wave front. Both the reduction of the local EGM amplitude and the frequency shift are uniquely associated with atrial lesion formation while recording from the ME electrodes.¹⁶ Furthermore, it has been shown that the focal EGM recordings from the ME uniquely provide high-fidelity localized recording allowing for the identification of small gaps in linear lesions and precise ablation targeting.¹⁶

Summary

This study demonstrated, for the first time, that focal EGM recordings from the ablation electrode provide a reliable method to define lesion maturation based on the maximal reduction of the EGM while maintaining efficacy and safety. The reduction of the recorded EGM activity from the ME and the increase in pacing threshold was associated with transmural atrial lesions and deep ventricular lesions, whereas, the 8-mm ring recordings provided inconsistent EGM reduction and pacing thresholds compared to the ME.

Limitations

The primary purpose of this investigation was to test the hypothesis that RF application time titration based on the maximal reduction of the ME EGM amplitude signifies atrial lesion maturation. Since the atrial tissues are thin they do not allow assessment of the maximal lesion thickness using this technique. We have applied the same RF power and RF application time based on the ME maximal EGM reduction to the ventricular tissues to define the maximal lesion size that can be accomplished with this technique. We did not apply higher power levels and longer RF application times to assess whether the lesions are larger and deeper, such investigation is currently underway. However, it has been previously shown that such a blinded approach will also increase the incidence of extracardiac injuries, steam pops, perforations, and char formation,¹⁷ which were not seen in this study. In addition, previous data show lesion dimensions cease to increase significantly after 20–30 s of RF ablation.^18,19

Significance

This work has demonstrated the utility of focused EGM monitoring to define atrial lesion maturation during RF ablation. Using this approach, RF delivery time is titrated to the ablation of the targeted tissues thereby avoiding expansion of the lesions to extracardiac structures.

References

1. Demazumder D, Mirotznik MS, Schwartzman D. Biophysics of radiofrequency ablation using an irrigated electrode. J Interven Cardiac Electrophysiol 2001; 5:377–389. [CrossRef] [PubMed]
2. Avitall B, Helms R, Koblish J, Sieben W, Kotov A, Gupta G. The creation of linear contiguous lesions in the atria with an expandable loop catheter. J Am Coll Cardiol 1999; 15:972–984. [CrossRef] [PubMed]
3. Sanchez JE, Kay NG, Benser ME, et al. Identification of transmural necrosis along a linear catheter ablation lesion during atrial fibrillation and sinus rhythm. J Interven Cardiac Electrophysiol 2003; 8:9–17. [CrossRef] [PubMed]
4. Jumrussirikul P, Atiga WL, Lardo A, et al. Prospective comparison of lesions created using a multipolar microcatheter ablation system with those created using pullback approach with standard radiofrequency ablation in the canine atrium. FACE 2000; 23:203–213. [CrossRef] [PubMed]
5. Peterson HH, Chen PX, Svendsen A, Haunso JH. Lesion dimensions during temperature-controlled radiofrequency catheter ablation of left ventricular porcine myocardium: impact of ablation site, electrode size, and convective cooling. Circulation 1999; 19:319–325. [CrossRef] [PubMed]
6. Nakagawa H, Yamanashi WS, Pitha JV, et al. Comparison of in vivo tissue temperature profile and lesion geometry for radiofrequency ablation with saline-irrigated electrode versus temperature control in canine thigh muscle preparation. Circulation 1995; 91:2264–2273. [CrossRef] [PubMed]
7. Avitall B, Hare J, Mughal K, Silverstein E. Segmented ablation electrode: a system for flexible lesion size. Circulation 1994; 90(Part 2, #671):1–126.
8. Avitall B. Split tip electrode catheter. WW Webster Jr, assignee. Patent 5,836,875. 1998. http://patents.justia.com/patent/5836875
9. Antz M, Otomo K, Nakagawa H, et al. Radiofrequency Catheter Ablation with the Split-Tip Electrode in the Temperature-Controlled Mode. Pacing Clin Electrophysiol 2001; 24:1765–73. [CrossRef] [PubMed]
10. Avitall B, Khan M, Krum D, Het al. The physics and engineering of transcatheter cardiac tissue ablation. J Am Coll Cardiol 1993; 22:921–932. [CrossRef] [PubMed]
11. Sjogren AL. Left ventricular wall thickness determined by ultrasound in 100 subjects without heart disease. Chest 1971; 60:341–346. [CrossRef] [PubMed]
12. Villamizar NR, Crow JH, Piacentino V, et al. Reproducibility of left atrial ablation with high-intensity focused ultrasound energy in a calf model. J Thorac Cardiovasc Surg 2010; 140:1381–1387. [CrossRef] [PubMed]
13. Scanavacca MI, D′Avila A, Parga J, Sosa E. Left atrial-esophageal fistula following radiofrequency catheter ablation of atrial fibrillation. J Cardiovasc Electrophysiol 2004; 15:960–962. [CrossRef] [PubMed]
14. Pappone C, Oral H, Santinelli V, et al. Atrio-esophageal fistula as a complication of percutaneous transcatheter ablation of atrial fibrillation. Circulation 2004; 109:2724–2726. [CrossRef] [PubMed]
15. Bunch TJ, Bruce KG, Johnson SB, et al. Analysis of catheter-tip (8-mm) and actual tissue temperatures achieved during radiofrequency ablation at the orifice of the pulmonary vein. Circulation 2004; 110:2988–2995. [CrossRef] [PubMed]
16. Price A, Leshen Z, Hansen J, et al. Novel Ablation catheter technology that improves mapping resolution and monitoring of lesion maturation. J Innov Cardiac Rhythm Manage 2012; 599–609. [CrossRef]
17. Dagres N, Hindricks G, Kottkamp H, et al. Complications of Atrial Fibrillation Ablation in a High-Volume Center in 1,000 Procedures: Still Cause for Concern? J Cardiovasc Electrophysiol 2009; 20:1014–1019. [CrossRef] [PubMed]
18. Avitall B, Mughal K, Hare J, et al. The effects of electrode-tissue contact on radiofrequency lesion generation. Pacing Clin Electrophysiol 1997; 20(Pt 1):2899–2910. [CrossRef] [PubMed]
19. Wittkampf F, Hauer R, Robles E. Control of radiofrequency lesion size by power regulation. Circulation 1989; 80:963–968. [CrossRef] [PubMed]