Author + information
- Received November 23, 2015
- Revision received February 5, 2016
- Accepted March 17, 2016
- Published online December 1, 2016.
- Duy T. Nguyen, MDa,
- Wendy S. Tzou, MDa,
- Matthew M. Zipse, MDa,
- Joshua D. Moss, MDb,
- Lijun Zheng, MSa and
- William H. Sauer, MDa,∗ ()
- aUniversity of Colorado, Section of Cardiac Electrophysiology, Division of Cardiology, Aurora, Colorado
- bUniversity of Chicago, Section of Cardiac Electrophysiology, Division of Cardiology, Chicago, Illinois
- ↵∗Reprint requests and correspondence:
Dr. William H. Sauer, Section of Cardiac Electrophysiology, University of Colorado, 12401 East 17th Avenue, B136, Aurora, Colorado 80045.
Objectives This study sought to determine whether partially insulated focused ablation (PIFA) catheters can minimize risk of injury to critical structures, such as the phrenic nerve and atrioventricular (AV) node, during ablation of adjacent myocardial tissue.
Background PIFA catheters using thermally conductive materials may have differential radiofrequency (RF) heating properties allowing for tailored RF application with more precision.
Methods Open-irrigated, 4- and 8-mm RF ablation catheter tips were insulated partially by coating one-half of their surfaces with a layer of vinyl, silicone, vinyl–silicone, polyurethane, or a composite of aluminum oxide/boron nitride (AOBN). These coated catheters or corresponding noninsulated catheters were positioned with 10 g of force on viable bovine myocardial tissue during RF application in an ex vivo setup. Tip temperatures, power, and lesion volumes were compared. The most effective coating, AOBN, was modified further by adding fenestrations to aid in passive cooling. PIFA catheters with fenestrated AOBN coating were then tested in an in vivo porcine model to target myocardial tissue adjacent to the AV node and the phrenic nerve.
Results PIFA catheters all demonstrated higher tip temperatures, although silicone- and AOBN-catheters demonstrated this to a lesser degree. Significant differences in lesion volumes and temperature-limited powers were noted between control, silicone, and AOBN tips. Steam pops were significantly higher for silicone but not AOBN. In contrast with non-PIFA catheters, injuries to the phrenic nerve and AV node during in vivo ablations with AOBN insulation positioned over these structures were reduced significantly.
Conclusions RF ablation using catheter tips partially coated with a thermally conductive insulation material such as AOBN results in larger ablation lesion volumes without temperature limitations. Partial insulation of the catheter tip will protect adjacent critical structures during RF ablation.
Effective radiofrequency (RF) ablation has been shown to be related to the size of the ablation catheter tip, temperature at the electrode–tissue interface, and duration and force of tissue contact (1–3). A limitation of RF ablation is the lack of durable lesions if only low powers, which may be safer, are delivered to myocardial tissue. In addition, anatomic considerations, such as proximity to critical structures including the atrioventricular (AV) node, phrenic nerve, or pericardium, can limit RF application. Furthermore, when ablating myocardial tissue, only 1 portion of the ablation catheter is in contact with the tissue being targeted; RF from the contralateral side of the catheter is either dispersed due to circulating blood or can unintentionally harm adjacent tissue.
We have modified existing catheters by covering a portion of the metallic tip with electrical insulation that is also thermally conductive. Insulating 1 side of an ablation catheter may allow for a more tailored cardiac ablation by improving heating under the noninsulated side of the catheter while decreasing undesired RF-mediated injury from the insulated side (4).
When insulation is applied partially to an ablation electrode, it can alter RF lesion geometry. Although insulation may result in higher tip temperatures, this can be minimized by using a larger catheter tip, by irrigating the catheter tip, and by adding fenestrations or vents in the insulation to allow for thermal release.
We sought to determine the most effective insulation material that will allow for asymmetrical lesion formation while protecting the tissue adjacent to the insulated aspect of the ablation catheter tip. We also ascertained the effects of “venting,” or interrupting the insulation to prevent tissue temperature limitations. Using what we found to be the most effective insulation, patterned with fenestrations to allow for additional cooling, we performed in vivo ablation near critical structures, such as the AV node and phrenic nerve, to determine the ability of PIFA to protect these structures.
Ex vivo model
Experimental protocols have been approved by the Institutional Animal Care and Use Committees of the University of Colorado and University of Chicago. An ex vivo model consisting of viable bovine myocardium, a submersible load cell, a circulating bath, and a deflectable sheath was assembled. A load cell was submersed in the bath and contained a section of viable bovine ventricular myocardium excised within 1 h of experimentation. This load cell measured force applied to the overlying myocardial tissue and was used to standardize application of energy. This ex vivo model has been validated and described in further detail elsewhere (5,6).
Catheter modification with electrical insulation using thermally conductive materials
A nonirrigated 4-mm RF catheter, nonirrigated 8-mm RF catheter, and an open-irrigated RF ablation catheter (Biosense-Webster, Diamond Bar, California) were insulated partially by coating one-half of their surfaces with a thin layer of thermally conductive material; the insulation materials tested included vinyl, silicone, vinyl–silicone, polyurethane, or a composite of aluminum oxide/boron nitride (AOBN). The coating suspensions were created with an epoxy vehicle and allowed to dry, leaving a thin (<0.1 mm) layer covering one-half of the metallic tip. For the externally irrigated catheter, the existing tip fenestrations were preserved, allowing for active cooling with saline irrigant on all sides of the catheter. These PIFA catheters or their corresponding noninsulated catheters (Biosense-Webster) were positioned with 10 g of force in a parallel position using a deflectable sheath (Agilis, St. Jude Medical, Secaucus, New Jersey). The insulation coating that was found to be most effective, AOBN, was further modified by adding vents, or fenestrations, into the coating to allow for thermal release/venting (Figure 1). Ablation with this vented AOBN was compared with standard AOBN coating, and temperatures were recorded.
Delivery of RF energy applied to myocardium
Using a temperature control mode, a series of RF lesions with each catheter was applied to recently excised bovine myocardium, with the noninsulated side parallel to and contacting the myocardium. Temperature limits for maximal power were set at 45°C for irrigated ablation and 55°C for 4- and 8-mm nonirrigated ablation. The number of lesions applied per ventricular section depended on the available endocardial surface. No lesions were placed over or in immediate proximity, defined as 5 mm, to papillary muscles or other lesions. Furthermore, no lesions were placed within 1 cm of section edge.
In vivo ablation of AV node and phrenic nerve
Yorkshire pigs (n = 12) were anesthetized and intravenous lidocaine (50 to 100 mg) or amiodarone (150 mg IV bolus followed by a 1 mg/min infusion) was used intraoperatively for prophylaxis of ventricular arrhythmias. Epicardial access was obtained under fluoroscopy using a 17-gauge Pajunk needle (Pajunk Medical Systems, Norcross, Georgia) and a 9-F sheath was placed in the epicardium. An electroanatomic map of the superior vena cava, right atrium, and epicardium was created using the CARTO3 mapping system (Biosense-Webster).
A decapolar catheter was used to pace and capture the right and left phrenic nerves, either in the endocardium or epicardium. A force-sensing PIFA irrigated tip catheter and a standard force-sensing irrigated tip catheter were used to deliver alternating PIFA and standard irrigated “control” ablation lesions directly below the site of phrenic nerve capture. After each ablation, pacing and capture of the phrenic nerve was moved inferiorly. Loss of phrenic nerve capture and ablation times were recorded and compared for PIFA and controls. During PIFA ablation, the insulated aspect of the catheter tip was oriented toward the phrenic nerve. Ablations were delivered at 50 W for 30 s with the same amount of force as measured by the force-sensing catheter (Thermocool SmartTouch, Biosense-Webster); ablation lesions were tagged by the electroanatomic mapping system. Saline irrigant was suctioned from the epicardium after each ablation.
For ablation near the AV node and His region, the bundle of His was mapped on the septal tricuspid annulus. Once the His region was annotated using electroanatomic mapping, either a standard irrigated catheter or a PIFA catheter was placed directly below the His and ablation was performed. AV block, time to AV block, and recovery of conduction after immediate cessation of ablation, if any, were recorded. During PIFA ablation, the insulated aspect of the catheter tip was oriented superiorly toward the His. Ablations were delivered at 50 W for 30 s with the same amount of force and direction of force vector as measured by the force-sensing catheter (Thermocool SmartTouch, Biosense-Webster); ablation lesions were tagged by the electroanatomic mapping system. The insulated portion of the catheter was visualized using the color-coded animated catheter display on the mapping system.
Ablation lesion volume measurements
Lesion volumes were acquired by analyzing tissue sections with a digital micrometer. Single lesion volumes were calculated using the equation for an oblate ellipsoid. For each lesion, maximum depth (A), maximum diameter (B), depth at maximum diameter (C), and lesion surface diameter (D) were measured.
Equation 1: volume of oblate ellipsoidwhere A is the maximum depth, B is the maximum diameter, C is the depth at maximum diameter, and D is the lesion surface diameter.
SPSS software (SPSS, Inc., Chicago, Illinois) was used to perform all calculations. Analysis of variance was used to compare continuous variables and the chi-square test was used for dichotomous comparisons in lesion characteristics from PIFA catheter-ablated myocardium versus lesions ablated by corresponding noninsulated catheters. The p values in Tables 1 and 2 are from paired comparisons within each condition versus the control. For comparisons using pooled results from the same animal or slab, hierarchal analysis with adjustment for possible bias due to clustering was performed.
Effect of partial insulation on tip temperatures and lesion sizes with RF energy delivery using temperature control mode
Significant differences in lesion volumes and peak temperatures were noted between control tip (untreated) and PIFA-catheter tips, using both low and high powers (Table 1). Partial insulation of an irrigated catheter tip with vinyl, vinyl–silicone, silicone, or polyurethane had significant tip temperature limitations at higher powers (50 W); the degree of limitation was less using cyanoacrylate and AOBN. In addition, steam pops occurred more often at 50 W for silicone and other insulation materials but not for AOBN. AOBN created larger lesions (210.9 mm3 vs. 126.9 mm3 for controls; p < 0.001), and did not cause higher rates of steam pops at 30 or 50 W (Table 1). Given the relative improvements in lesion size and potential safety compared with the other PIFA types, AOBN then was selected for further testing and development.
Effect of partial insulation on targeted myocardial tissue temperature dispersion
During a 60-s ablation at 20 W (power control mode), the mean temperatures recorded at 3- and 5-mm depths beneath the irrigated ablation catheter tip were significantly higher with AOBN-PIFA compared with ablation using a standard 4-mm catheter (Figure 2). Further modification of the AOBN partial insulation with vents (Figure 1) improved catheter tip–tissue interface temperatures (Figure 2), thereby overcoming temperature limitations and decreasing steam pops. Vented AOBN achieved maximum 50 W powers with larger lesion volumes compared with control ablations (155.5 vs. 123.6 mm3; p < 0.001), but smaller lesions than nonvented AOBN insulated ablation (Table 2).
Protection of bundle of His and AV node during RF ablation with an externally irrigated PIFA catheter in an in vivo porcine model
Ablation of the AV junction region, defined as the area bordered by the coronary sinus and tricuspid valve and below the level of the bundle of His, was performed in 12 pigs (6 with control and 6 with PIFA catheters). Application of RF energy was purposely adjacent to the His region (Figure 3). Ablation near the His region with an open irrigated, vented AOBN-insulated PIFA catheter did not result in heart block (Figure 3) (n = 0/6) compared with heart block occurring in all controls ablated with a nonmodified irrigated catheter (n = 6/6; p = 0.002).
Protection of phrenic nerve during RF ablation with an externally irrigated PIFA catheter in an in vivo porcine model
Using a porcine model, we evaluated the effect of ablation near the phrenic nerve using a partially insulated catheter tip by purposely ablating on or near the phrenic nerve on the endocardial and epicardial surfaces of the heart. We used a decapolar catheter to identify sites of right and left phrenic nerve capture on endocardial and epicardial surfaces. While stimulating the phrenic nerve superiorly to sites of ablation, we alternated vented AOBN-insulated PIFA with control, irrigated RF applications on both endocardium and epicardium near the right and left phrenic nerves (Figure 4). Ablations were performed at 50 W for 30 s or were halted immediately if there was loss of phrenic nerve capture during RF application.
In 7 animals, endocardial control ablations near the left and right phrenic nerves using control, irrigated-tip ablation catheters caused loss of phrenic nerve capture in 51 of 53 ablations, compared with 10 of 50 ablations for PIFA catheters. The average time to loss of capture during 50 W RF ablation was 24 ± 9 s for PIFA, compared with 8 ± 8 s for control (p < 0.001). In 3 animals, epicardial ablations resulted in a lower incidence of distal phrenic nerve paralysis compared with controls (7 of 18 ablations vs. 22 of 23 ablations). In affected animals, the average time to loss of phrenic nerve capture during epicardial ablation was 21 ± 12 s for PIFA compared with 5 ± 5 s for control (p < 0.001).
This study has shown that, by partially insulating 1 aspect of the ablation catheter, RF current density can be increased toward the noninsulated side and thereby create larger and asymmetric ablation lesions. In this study, we evaluated a variety of insulation materials that can be used to partially insulate ablation catheters. Insulation with a composite of AOBN seemed to be the most effective thermally conductive electrical insulation with the least amount of observed temperature limitations during high-power ablation. An additional modification was performed by creating small vents in the insulation to allow for enhanced thermal release with a decrease in current density applied to the tip–tissue interface using standard powers. This modification led to improved cooling at the noninsulated catheter tip–tissue interface and also decreased risks of steam pops during ablation at high powers. Using this AOBN-insulated and -vented catheter, we demonstrated that ablation near the AV node did not cause heart block, and ablation near the phrenic nerve had less risk of injury compared with standard noninsulated RF ablation catheters.
Partial insulation and RF ablation
We previously demonstrated that partial insulation of the ablation catheter tip increases the effective current density on the noninsulated surface, which leads to larger asymmetric lesions (4). To allow for improved thermal cooling and to mitigate temperature limitations at the tissue interface with the noninsulated side of the catheter tip, we interrupted the insulation covering with fenestrations. The effective increase in surface area of RF energy dispersion resulted in a slightly attenuated RF current density concentration at the tip–tissue interface. However, although this catheter tip design creates smaller ablation lesions compared with nonfenestrated PIFA catheters, the safety profile is improved, due to increased convective cooling and lower risk of temperature limitations and coagulum formation. This fenestrated design still carries the benefit of partial insulation and shielding of adjacent tissue, because the exposed metal is within a recess created by the fenestrated insulation, thus preventing contact with nontargeted tissue. Furthermore, catheters with fenestrated insulation still create asymmetric lesions that are larger than those with noninsulated catheters.
Our prior studies have shown that ablation with an insulated catheter causes tissue necrosis only inferior to the top border of the catheter, unlike noninsulated standard catheters. Because of this altered distribution of energy, tissues immediately adjacent to the insulated superior aspect of the catheter remained protected, as was observed in the present in vivo porcine model using this catheter during ablation adjacent to the AV node and phrenic nerve. This phenomenon was most effective when the catheter could be placed such that the insulated aspect of the catheter tip was fully oriented toward the structure to be shielded from ablation, and none of the noninsulated side was directed toward it. For instance, AV nodal conduction was not affected during PIFA ablation, whereas control ablations resulted in complete heart block.
In contrast, ideal PIFA catheter orientation was more difficult to achieve during both endocardial and epicardial ablation near the phrenic nerve. It was more difficult to orient the catheter such that the entire insulated aspect of the catheter was facing the phrenic nerve without having some of the noninsulated, distal portion of the catheter tip also adjacent to the structure. This difficulty likely explains why we saw a higher rate of phrenic nerve injury with PIFA compared with AV node injury, even though the rates were still lower compared with control ablations. Improved catheter orientation and incorporation of contact force and vector may complement the use of insulated catheters (7,8). Further modifications to the design of insulation patterns covering a RF ablation catheter tip may also improve its safety profile.
Potential clinical applications for partial insulation using thermally conductive materials
RF catheter ablation for the treatment of arrhythmias commonly involves RF application near the AV node, phrenic nerve, and on the epicardial surface opposite the parietal pericardium. Ablation near these structures can lead to complications, including heart block, phrenic nerve paralysis, and pericarditis (9–13). Although all current RF ablation catheters are radially symmetric and deliver RF circumferentially, RF energy delivered asymmetrically from a single side of the catheter may improve the safety of catheter ablation when there is a risk of unintended collateral injury to vital structures such as these. Safety may also be improved because, with RF energy concentrated to a single side of the catheter, effective ablation can be achieved at lower powers. Concentrated asymmetric RF delivery results in deeper myocardial lesion formation, and may therefore have additional applications for targeting the interventricular septum, papillary muscles, midmyocardial circuits, and thick cavotricuspid isthmuses (14–16).
Partial insulation and asymmetric RF delivery can be achieved using a thin coating of thermally conductive material on one-half the surface of a catheter tip to direct RF energy preferentially to the noninsulated side of the catheter tip. Previous ex vivo studies of the PIFA catheter have shown that the noninsulated side delivers more tissue heating with less power, resulting in larger lesions whereas the insulated side has decreased temperature changes and less tissue injury (4). This in vivo study demonstrates that this catheter design and resultant asymmetric RF delivery improves the safety of catheter ablation when delivering energy adjacent to critical structures.
The limitations of ex vivo studies have been detailed previously (4,17–19) and include variability in circulating bath currents, catheter contact or angulation, passive catheter cooling, and presence of ischemic myocardium due to nonperfusion. These variables were nondifferential among controls and test groups, and repeated measurements within groups were performed to reduce the impact of these variables.
For in vivo animal studies, assessment of injury to the phrenic nerve was performed by delivering lesions from the PIFA and noninsulated ablation catheters in an alternating fashion, with multiple RF applications used for data analysis from each pig. One unavoidable limitation of this technique was that RF could not be delivered to the exact same anatomic site from both catheters for comparison. Instead, immediately adjacent sites were used for comparison. This limitation is unlikely to affect our results, however, as the phrenic nerve is not a single discreet point structure; rather, it has sufficient bordering myocardial tissue to allow for multiple lesion applications. Slight anatomic variation would affect lesions from both catheters equally when averaged across all delivered lesions and would tend to bias our results toward the null. Operators were not blinded to the ablation technology; thus, we established objective measures of ablation targets (His signals and phrenic capture) and objective measures of ablation efficacy, including heart block or loss of phrenic capture, to decrease the risk of subjectivity.
There may be some concern that the application of the nonsymmetrical catheter design could lead to inadvertent placement of the insulated side of a PIFA catheter tip directed toward the intended myocardial tissue and the noninsulated side toward the unintended target, with deleterious consequences affecting clinical results and safety. In addition, the measured contact force may not be equal depending on which direction it is being applied. We believe that the use of electroanatomic mapping, fluoroscopic imaging, intracardiac echocardiography, and contact and vector forces would minimize the risk of this possibility, and we did not see this as a significant limitation when using these tools in conjunction with a bidirectional, deflectable catheter wherein the orientation could be relatively easily determined. The ideal target tissue with this technology will likely be in those cases where parallel (to target tissue) ablation is performed to harness both the safety and efficacy advantages; these advantages may be negated when the PIFA catheter is perpendicular to the target tissue.
Last, although our results are intriguing and have implications for clinically relevant ablation strategies, further studies involving partially insulated catheters are needed with further animal studies and clinical trials in humans to confirm both the safety and efficacy of this design. Furthermore, for this proof-of-principle study, we used an irrigated catheter and ablated at the AV node/fast pathway region, near the bundle of His, to achieve our maximal endpoints. However, in clinical practice, these strategies are seldom performed.
In our ex vivo model, we have refined and identified an effective insulation material and pattern that balances efficacy with safety. This composite of aluminum oxide and boron nitrite insulates against unintended RF application while concentrating it toward the noninsulated aspect of the catheter. Tissue temperature limitations can be mitigated through the use of irrigation, lower powers, and insulation vents. Adjacent tissues exposed to the insulated surfaces of ablation catheters had lower risks of injury, as demonstrated during ablation near the AV node and phrenic nerve. The use of partially insulated catheters may have a clinical role for improving efficacy or reducing the risk of complications during ablation of some arrhythmias.
COMPETENCY IN MEDICAL KNOWLEDGE: Catheter ablation for the treatment of cardiac arrhythmias is associated with procedural risk, which includes collateral damage to critical anatomic structures adjacent to targeted tissue, such as the AV node and phrenic nerve. A partially insulated focused ablation (PIFA) catheter, in which the catheter tip is partially coated with thermally conductive insulation material, may help protect these structures while preferentially delivering radiofrequency to tissue oriented adjacent to the noninsulated tip, thereby improving the safety and efficacy of catheter ablation.
TRANSLATIONAL OUTLOOK: The results of this in vivo porcine study in which the PIFA catheter limited injury to the phrenic nerve and AV node when ablating near these structures are encouraging, although larger scale studies are needed to prove safety and efficacy in humans which would translate to clinical practice.
For a supplemental figure, please see the online version of this article.
Drs. Sauer and Nguyen have received significant research grants from Biosense Webster and CardioNXT; and educational grants from St Jude Medical, Boston Scientific, and Medtronic. Drs. Sauer and Nguyen have a provisional patent on partially insulated focused catheter ablation. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- aluminum oxide/boron nitride
- partially insulated focused ablation
- Received November 23, 2015.
- Revision received February 5, 2016.
- Accepted March 17, 2016.
- American College of Cardiology Foundation
- Haines D.E.,
- Verow A.F.
- Jesel L.,
- Sacher F.,
- Komatsu Y.,
- et al.
- Aryana A.,
- d'Avila A.
- Hintringer F.,
- Hartikainen J.,
- Davies D.W.,
- et al.
- Yoshida K.,
- Yokokawa M.,
- Desjardins B.,
- et al.
- Yamada T.,
- Doppalapudi H.,
- McElderry H.T.,
- et al.
- Nguyen D.T.,
- Barham W.,
- Moss J.,
- et al.