ORIGINAL RESEARCH published: 12 November 2021 doi: 10.3389/fmedt.2021.747609 Systematic Characterization of High-Power Short-Duration Ablation: Insight From an Advanced Virtual Model Argyrios Petras 1† , Zoraida Moreno Weidmann 2† , Massimiliano Leoni 1 , Luca Gerardo-Giorda 1,3‡ and Jose M. Guerra 2*‡ 1 Johann Radon Institute for Computational and Applied Mathematics (RICAM), Austrian Academy of Sciences, Linz, Austria, 2 Department of Cardiology, Hospital de la Santa Creu i Sant Pau, Centro de Investigación Biomédica en Red Enfermedades Cardiovaculares (CIBERCV), Universitat Autónoma de Barcelona, Barcelona, Spain, 3 Institute for Mathematical Methods in Medicine and Data-Based Modelling, Johannes Kepler University, Linz, Austria Edited by: Peter Edward McHugh, Background: High-power short-duration (HPSD) recently emerged as a new approach National University of Ireland Galway, Ireland to radiofrequency (RF) catheter ablation. However, basic and clinical data supporting its Reviewed by: effectiveness and safety is still scarce. Frederic Heim, Objective: We aim to characterize HPSD with an advanced virtual model, able to Université de Haute-Alsace, France Wentao Feng, assess lesion dimensions and complications in multiple conditions and compare it to Beihang University, China standard protocols. *Correspondence: Methods: We evaluate, on both atrium and ventricle, three HPSD protocols (70 W/8 s, Jose M. Guerra jguerra@secardiologia.es 80 W/6 s, and 90 W/4 s) through a realistic 3D computational model of power-controlled † These authors have contributed RF ablation, varying catheter tip design (spherical/cylindrical), contact force (CF), blood equally to this work and share first flow, and saline irrigation. Lesions are defined by the 50◦ C isotherm contour. Ablations authorship are deemed safe or complicated by pop (tissue temperature >97◦ C) or charring (blood ‡ These authors share last authorship temperature >80◦ C). We compared HPSD with standards protocols (30–40 W/30 s). We analyzed the effect of a second HPSD application. Specialty section: This article was submitted to Results: We simulated 432 applications. Most (79%) associated a complication, Cardiovascular Medtech, especially in the atrium. The three HPSD protocols performed similarly in the atrium, while a section of the journal Frontiers in Medical Technology 90 W/4 s appeared the safest in the ventricle. Low irrigation rate led frequently to charring Received: 26 July 2021 (72%). High-power short-duration lesions were 40–60% shallower and smaller in volume Accepted: 05 October 2021 compared to standards, although featuring similar width. A second HPSD application Published: 12 November 2021 increased lesions to a size comparable to standards. Citation: Petras A, Moreno Weidmann Z, Conclusion: High-power short-duration lesions are smaller in volume and more Leoni M, Gerardo-Giorda L and superficial than standards but comparable in width, which can be advantageous in the Guerra JM (2021) Systematic Characterization of High-Power atrium. A second application can produce lesions similar to standards in a shorter time. Short-Duration Ablation: Insight From Despite its narrow safety margin, HPSD seems a valuable new clinical approach. an Advanced Virtual Model. Front. Med. Technol. 3:747609. Keywords: radiofrequency ablation, high-power short-duration, computational modeling, lesion science, ablation doi: 10.3389/fmedt.2021.747609 catheter Frontiers in Medical Technology | www.frontiersin.org 1 November 2021 | Volume 3 | Article 747609 Petras et al. Virtual Characterization of HPSD Ablation INTRODUCTION comparison of the different ablation settings. The mathematical model, based on an in-vitro experimental setup (12), is built Radiofrequency (RF) catheter ablation is the cornerstone within a three-dimensional computational framework described treatment for cardiac arrhythmias. Radiofrequency lesions previously by our group (11). It considers tissue elasticity, depend mainly on the power delivered to the tissue, catheter thermoelectric interaction between electrode and ablation area, stability, and contact force (CF) over time (1, 2). In the and convective cooling by blood flow and irrigation saline. In initial development of RF ablation in the 90’s, a power Petras et al. (11) the model has also been validated against in- titration was performed according to efficacy and safety targets vitro experimental data for standard RF protocols. A detailed (temperature registered by the catheter, impedance drop, signal description of the model is given in the Supplementary Material. analysis), between 20 and 50 W. Clinical experience and extensive The biophysical properties of the tissue introduced in the research in the field, aiming at maximizing the lesion size and model are tissue density, thermal conductivity and heat capacity minimizing the risk of complications, led to protocols where (Supplementary Table 1 of the Supplementary Material), the power and duration have been standardized according to whose values are based on experimental data registered in IT’IS the arrhythmogenic substrate. For 4 mm-tip catheter the general database (13). Heat expansion through the tissue occurs in a consensus is to limit the maximum power to 30–50 W over 30– radial manner. We assume a linear temperature-dependent 90 s (2, 3). Nevertheless, catheter stability and optimal CF over decrease of specific heat (0.42 per 1◦ C increase) and thermal time are challenging to keep in a beating heart. This may result conductivity (0.05% per 1◦ C increase), and a linear temperature- in smaller and inconsistent lesions that hamper the creation of dependent increase in electrical conductivity (0.15% per 1◦ C effective lines of electrical block and prolong the duration of increase), following the experimental studies of Haines and the procedures. Watson (14) and Rossmanna and Haemmerich (15). Tissue Aimed at overcoming these limitations of conventional heterogeneity may influence the biophysical properties, however, protocols, a new RF ablation paradigm has emerged in the last due to the lack of data in this respect, we neglected this feature in years. Known as high-power short-duration (HPSD) ablation, the model. The governing equation for the thermal problem is the these protocols are based on increasing maximum power Bioheat Equation, while the blood flow and its interaction with and markedly shortening the application. A heterogeneity of the irrigation saline in the blood compartment are modeled using power/duration settings, from 35 to 60 W and 6 to 15 s, have the incompressible Navier-Stokes equations for fluid dynamics been proposed and tested for pulmonary vein isolation (4). Some (16). The model is solved by finite elements through our self- groups have investigated more aggressive protocols with power developed software in FEniCS-HPC (www.fenics-hpc.org). The ranging from 70 to 90 W during 4–8 s (5, 6). However, lesion computational geometry features about 5,000,000 tetrahedra science of HPSD is largely unknown and the results of in-vivo with a maximum element size of 1 mm and a minimum of and in-vitro studies comparing these lesions to standard ones 0.01 mm in the area of the electrode, where higher computational are contradictory (6–8). Moreover, clinical data supporting their accuracy is needed. Simulations were carried out on the cluster effectiveness and safety is limited (9, 10). Additionally, as most of Hipatia of BCAM (the Basque Center for Applied Mathematics, the clinical studies are on pulmonary vein isolation, experience Bilbao, Spain) which features 18 nodes (1 with Nvidia Tesla on other atrial substrates is limited and absent on the ventricle. K40 GPU) for 624 cores with 4 TB RAM and Infiniband The development of advanced virtual models for RF ablation network connectivity. allows to simulate lesion formation mimicking infinite scenarios. A systematic and realistic evaluation of RF ablation can be performed by changing parameters as catheter type and shape, Electrode Tip Design ablation protocol, blood and irrigation flow, CF, or cardiac We consider two open-irrigated electrode designs used in chamber. This will grant an understanding of RF lesion formation clinical practice, and also used in a previous experimental study and its complications beyond that obtained through classical performed by our group (12), one with a hemispherical tip and in-vitro and in-vivo assessment. This novel research approach one with a cylindrical tip (7 French, 3.5 mm length). Both feature proves of particular interest in early stages of new ablation six irrigation pores (0.5 mm diameter) (Figure 1). strategies like HPSD. We aim to systematically characterize different HPSD ablation Lesion Assessment protocols by means of an advanced virtual model developed At 50◦ C, cardiomyocytes exhibit a permanent loss of excitability, by our group (11). Lesion dimensions and complications will as well as a thermally induced necrosis leading to irreversible be evaluated and compared to those of standard conventional myocardial injury, which is the target of RF ablation. In protocols, providing valuable insight in view of the clinical light of the above consideration, virtual lesions are identified application of HPSD. by the isotherm contour of 50◦ C within the cardiac tissue, considering irreversible tissue damage at 50◦ C (17, 18). METHODS The assessment of the lesion dimensions is performed with ParaView (19). We measure lesion volume, maximum width Virtual Model (W), maximum depth (D), and depth at the maximum width Our computational model of RF catheter ablation simulates a (DW) (Figure 1), considering the zero at the undeformed human heart considering fixed conditions, to standardize the endocardial surface. Frontiers in Medical Technology | www.frontiersin.org 2 November 2021 | Volume 3 | Article 747609 Petras et al. Virtual Characterization of HPSD Ablation FIGURE 1 | Virtual lesion created with the two considered catheters. (A) Spherical tip and (B) Cylindrical tip. The area of irreversible tissue lesion (tissue temperature ≥50◦ C) is delimited by the blue line. Measurements: maximum depth (D), depth at the maximum diameter (DW), maximum diameter (W). Complications Assessment TABLE 1 | Summary of the different ablation settings and physiological scenarios A steam pop is considered to occur when tissue temperature considered combining catheter tip, contact force, power, duration, irrigation rate, blood flow, and cardiac chamber. ≥97◦ C. We use a temperature threshold <100◦ C to account for the natural fluctuation of water boiling point depending on Ablation settings Catheter tip Spherical tissue composition and other physical conditions (20). Although Cylindrical reaching a boiling temperature doesn’t automatically result in Contact force 5g pop, given the seriousness of this complication, we equate the 10 g 15 g risk of pop with the occurrence of it. Charring is assumed to be 20 g initiated at blood temperature of 80◦ C. The occurrence of one Power/Duration 70 W/8 s of the two complications ends the application, precluding the 80 W/6 s development of the other. Applications that finish without any 90 W/4 s complication are deemed safe. Irrigation rate 17 ml/min 30 ml/min 60 ml/min Ablation Settings Physiological Blood-flow 0.0 m/s Three different HPSD ablation protocols are used (70 W/8 s, scenarios 0.1 m/s 80 W/6 s, 90 W/4 s) delivered in power-control mode without 0.5 m/s temperature limit. The model considers a fixed impedance of 120 Ablation site Atrium Ohms, mimicking the behavior of contemporary RF generators, Ventricle that vary the RF current delivery according to impedance changes, to ensure the delivery of the selected power. Since the chosen power is a fixed parameter, changes in baseline impedance and during ablation are disregarded. Each protocol is simulated high rate (17, 30, and 60 ml/min, respectively, according to the on the atrium and the ventricle. All virtual applications are technical specifications of the commercial catheters). performed perpendicularly to the endocardium. Based on clinical All simulated combinations are summarized in Table 1. data, we assume similar biophysical properties for both cardiac regions, only differenced by thickness (6 mm in atrium, 12 mm in Design of the Study ventricle) (21). Four increasing CF from 5 to 20 g are simulated The study is structured in three steps: resulting in progressive deformation of the tissue. In order to simulate representative cardiac regions (22) three A. We evaluate the safety of the different HPSD ablation settings blood flow velocities are used: no-blood flow (0 m/s), as expected and protocols. in epicardium or below a valve leaflet; low-blood flow (0.1 m/s), B. We assess the efficacy of HPSD protocols identified as the as expected in healthy atrium or pulmonary veins; and high- safest in the previous step. For both chambers, we compute blood flow (0.5 m/s), as expected at the cavo-tricuspid annulus, HPSD lesion sizes and compare them with those generated ventricle, or outflow tracts. by standard applications according to the current standard- Additionally, three different saline catheter irrigation rates, of-care (30 W/30 s for the atrium and 30–40 W/30 s for the are simulated for each HPSD protocol: normal, high, and very ventricle, CF 10 g, irrigation rate 17 ml/min) (3, 23). Frontiers in Medical Technology | www.frontiersin.org 3 November 2021 | Volume 3 | Article 747609 Petras et al. Virtual Characterization of HPSD Ablation FIGURE 2 | Safety outcomes of HPSD protocols according to the different parameters in Table 1. HBF, high blood-flow (0.5 m/s); LBF, low blood-flow (0.1 m/s); NBF, no blood-flow (0.0 m/s). Frontiers in Medical Technology | www.frontiersin.org 4 November 2021 | Volume 3 | Article 747609 Petras et al. Virtual Characterization of HPSD Ablation C. We analyze the impact of a second HPSD application on top Safety of HPSD of the first one. We evaluate changes in terms of safety and The main findings are summarized in Figure 2 and lesion size at increasing time intervals (2, 4, 6, 8, and 10 s) Supplementary Table 1. Five conclusions can be highlighted: between the first and second application. Irrigation rate is kept a. The safety margin of HPSD ablation is, in general, tight, at 2 ml/min between the two applications, while the protocol considering that several of the applications associate some irrigation rate is restored 1 s prior to the second application. complications [342/432 (79%)]. This finding is more notable Biophysical tissue characteristics for the second application in the atrium than in the ventricle [199/216 (92%) vs. account for the changes in thermal and electrical conductivity 143/216 (66%)]. resulting from the first application. b. Limited electrode tip cooling, through saline flow and/or For B and C, we simulate a low-blood flow (0.1 m/s) for blood flow, mostly leads to charring: at 17 ml/min, 94% atrium and ventricle, mimicking in the latter the less favorable of applications with no-blood flow (45/48) lead to charring conditions for ablation and prone for complications, as those in contrast with 15% with high-blood flow (7/48). Charring that can be found in areas between papillary muscles and largely appears at the very beginning of the application (<1 s). cardiac wall. Increasing the flushing rate avoids charring and allows to continue the application, which is however often interrupted shortly before the end because of the appearance of pop, as it RESULTS is the case in 67% of applications at 60 ml/min and no-blood flow (32/48). A total of 432 RF ablation lesions are simulated by c. At low CF, all HPSD protocols exhibit low incidence of pop. applying the different predefined ablation parameters In the specific, pops occur in 31% of applications in the atrium (catheter tip design, CF, power, duration, saline (17/54), and never (0/54) in the ventricle. Higher CF result safe irrigation rate) and physiological scenarios (blood flow, only in the ventricle, where 41% applications (66/162) are free ablation site). of pop for CF >5 g. FIGURE 3 | Overall safety comparison of the two different catheter tip designs. Frontiers in Medical Technology | www.frontiersin.org 5 November 2021 | Volume 3 | Article 747609 Petras et al. Virtual Characterization of HPSD Ablation d. Different catheter tip designs show varied behaviors Efficacy of HPSD (Figure 3). Spherical tips appear relatively safe in both In order to assess the efficacy of HPSD protocols, we considered chambers, while cylindrical tips only when used in the the safest setting emerged from the previous set. These are ventricle. Nevertheless, cylindrical tips in the ventricle allow identified from Figure 2 as the configurations that ensure a safe the use of larger CF. termination for all protocols. These settings are represented by e. The complication distribution (charring vs. pop) is similar CF 5 g, irrigation rate 60 ml/min, using a spherical tip in the for both catheters in the ventricle, whereas in the atrium the atrium and a cylindrical tip in the ventricle. spherical tip produces more charring and the cylindrical more The analysis of the different HPSD protocols shows that as pops (Figure 3). power increases and duration diminishes, lesions tend to be FIGURE 4 | Comparison between HPSD and standard ablation protocols in atrium (A) and ventricle (B). Radar charts show the percentage comparison against a reference standard protocol (30 W/30 s in the atrium, 40 W/30 s in the ventricle). The associated tables present the absolute values. D, lesion depth; DW, Depth at maximum width; V, volume; W, lesion width. Frontiers in Medical Technology | www.frontiersin.org 6 November 2021 | Volume 3 | Article 747609 Petras et al. Virtual Characterization of HPSD Ablation narrower and shallower, resulting in a smaller volume (Figure 4). protocols, while in the ventricle HPSD lesions are smaller High-power short-duration generates lesions that are always when compared to the ones of the 40 W/30 s protocol, but more superficial than those obtained with standard protocols similar to the 30 W/30 s standard ones (with the exception but comparable in width. In the atrium HPSD lesions are of the 90 W/4 s protocol whose lesions were consistently of always smaller in volume than those obtained with standard smaller volume). FIGURE 5 | Lesion comparison between 90 W/4 s and standard protocols in the ventricle at different CF. The table collects the absolute values. (A) Comparison of 90 W/4 s vs. 30 W/30 s protocol at CF 5 and 20 g; (B) comparison of 90 W/4 s vs. 40 W/30 s protocol at CF 5 and 20 g. Frontiers in Medical Technology | www.frontiersin.org 7 November 2021 | Volume 3 | Article 747609 Petras et al. Virtual Characterization of HPSD Ablation Additionally, we analyze the effect of increasing CF in the tendency to slightly underestimate width. This is likely related ventricle for the 90 W/4 s protocol with a cylindrical catheter to an anisotropic electrical/thermal conduction in the tissue. which was the only HPSD setting that allowed safe applications Nevertheless, since this anisotropic conduction has never been at higher CF. At CF 20 g, lesions become comparable to the evaluated neither in-vitro nor in-vivo, we had to use an isotropic ones from standard 30 W/30 s protocols but remain markedly conduction for our model. smaller in volume than those from 40 W/30 s protocols (Figure 5; The study of Kotadia et al. (24) collects experimental data Supplementary Figure 1). on a variety of tissues, featuring several different protocols, both standard and HPSD. Unfortunately, several details such as CF Repeated HPSD Applications and saline irrigation for the studies considered in their review A second identical HPSD application on top of a first one is are incomplete. Moreover, as we observed in a previous study, simulated after a variable time interval. We consider a spherical the interspecies variability in biophysical properties results in tip in the atrium and a cylindrical tip in the ventricle, CF 5 g, low- non-negligible differences in the lesion size (25). Hence, a direct blood flow and irrigation rate of 60 ml/min, in accordance with validation of our HPSD result is not really feasible. Nevertheless, the results of the first step. In general, all lesions increase in size we can observe (Table 2) an adequate correlation with depth and after a repeated application (Figure 6; Supplementary Table 2). a slight underestimation of width, for both HPSD and standard As the time interval between applications gets longer, lesion size protocols. Such consistency in width underestimation does not increase is reduced. Furthermore, as the interval decreases, the hamper a fair comparison between our results for standard and risk of pop appearance is higher. For both cardiac chambers, the HPSD protocols. 80 W/6 s protocol shows the worst profile, requiring the longest interval between both applications to avoid complications. Safety Margin of HPSD In the atrium, increase in width is more pronounced than The experience with standard RF protocols is wide and the in depth, being more evident at 90 W/4 s. Only 70 W/8 s currently used have demonstrated to be mostly safe (2, 3). applications with a 4-s interval reach a final lesion volume that High-power short-duration represents a new approach to RF exceeds that of a lesion performed with a standard protocol ablation aimed at reducing ablation time and overcoming (Supplementary Figure 2). some limitations of standard protocols. In agreement with In the ventricle, the lesion increases similarly in depth and the published data (9, 10), our results show that HPSD can width. The best trade-off between safety and lesion increase is be effectively used, although its safety margin is narrow. reached with a 90 W/4 s protocol and 6-s interval (showing a According to our study, the main determinants of complications 77% volume increase). The final lesion after a second application are the irrigation rate and the catheter tip design, the latter is larger than the standard 30 W/30 s lesion for any HPSD mostly related to the chamber where RF is applied. To the protocol, especially in terms of width, but generally of smaller best of our knowledge, this is the first time that an association volume when compared to those obtained with 40 W/30 s between catheter tip shape, anatomical location of ablation (Supplementary Figure 4). and safety is shown. These findings suggest that HPSD should be clinically used with caution, additional technological safety DISCUSSION tools should be implemented in the catheters and generators, and operators should use the technology strictly following This is the first study where an advanced virtual model is used to industry recommendations ensuring stable position and systematically evaluate HPSD ablation in atrium and ventricle. In adequate CF. addition, our innovative model provides a comprehensive insight in the mechanisms of RF ablation, setting the basis for future Influence of Catheter Tip Design research in the field. The main advantage of a computer model Catheter tip shape has a crucial impact on the safety profile of is that it enables a controlled introduction of a wide range of HPSD. We observed that while a spherical catheter tip provides parameters that influence lesion formation and produces direct the best trade-off between safety and efficacy in the atrium, the results. It circumvents the need for extensive statistical analysis opposite occurs in the ventricle. required by in-vivo and in-vitro experimentation where such A possible explanation can be found in the physics of parameters are not always under control. RF ablation. The power delivered to the tissue increases The major findings of the study are: 1. The safety margin of proportionally to the percentage of the catheter tip that is in HPSD is in general tight; 2. Catheter tip design and irrigation contact with the endocardium (1), which is different for each rate are critical to ensure safe and efficient applications; 3. High- catheter according to the CF. At low CF, the entire base of power short-duration lesions are in general shallower and smaller the cylindrical tip is in contact with the tissue, while only a in volume compared to standard lesions; 4. A second HPSD tiny portion of the spherical tip does so. Increasing the CF, the application after a suitable time lag increases lesion size. tissue progressively deforms, and the contact surface increases with the pressure. This increase is more pronounced for the Model Validation spherical tip. As a consequence, at low CF (<10 g) spherical tips Our model has been previously validated with standard in-vitro feature a smaller contact surface between tissue and electrode that ablation protocols of porcine tissue (11). For this setting, we cylindrical tips, while as the CF increases (>10 g) this relationship have shown that our lesions correlate well in depth but have a is reversed (Supplementary Figure 3). As a result, at low CF the Frontiers in Medical Technology | www.frontiersin.org 8 November 2021 | Volume 3 | Article 747609 Petras et al. Virtual Characterization of HPSD Ablation FIGURE 6 | Effect on lesion size from a second HPSD application on top of a first one as a function of the time interval between the two. Ablations are performed with spherical tip in the atrium and cylindrical tip in the ventricle. The percentage increase in depth, width and volume is represented as bar charts for atrium (A) and ventricle (B). Striped bars indicate a pop occurrence during the second application. spherical tip delivers less power than the cylindrical one while Although few studies have analyzed the role of at high CF the opposite occurs. Heat dissipation in the thinner electrode tip shape on standard ablations, a similar atrial wall is lower (14), limiting the power that can be safely pattern was already observed with standard protocols in delivered to this chamber. In this case, only the spherical tip at previous in-vitro studies (12) comparing two commercially low CF appears to remain within the safety margins. Contrarily, available catheters, ThermoCool R (Biosense Webster, the thicker ventricular wall allows higher power delivery, and the cylindrical design) and CoolPath R (St Jude Medical, cylindrical tip appears more effective and safer. spherical design). Frontiers in Medical Technology | www.frontiersin.org 9 November 2021 | Volume 3 | Article 747609 Petras et al. Virtual Characterization of HPSD Ablation TABLE 2 | Comparison between HPSD and standard ablation protocols derived from several in-silico, ex-vivo, and in-vivo studies compiled by Kotadia et al. (24) against the lesions simulated by our model in atrium and ventricle. Protocol Energy Depth Width (W/s) (J) (mm) (mm) Kotadia et al. Virtual lesion Kotadia et al. Virtual lesion Atrium Ventricle Atrium Ventricle 30 W/30 s 900 5.1–5.7 5.31 4.53 7.9–8.9 7.59 6.49 70 W/8 s 560 4.0–4.3 3.47 3.68 10.5–10.8 6.82 7.10 80 W/6 s 480 2.9–3.9 3.29 3.49 7.9–11.2 6.96 6.96 90 W/4 s 369 2.6–3.6 2.97 3.07 5.9–10.3 6.20 6.56 For that purpose, the power/s protocols are converted to total energy delivery. D, lesion depth; W, lesion width. Catheter Tip Design and Effective Cooling standard applications. However, this observation could be related As previously demonstrated (26), the use of catheter cooling to other factors like catheter stability or constant CF during systems allows larger power delivery preventing charring, application, more difficult to maintain during standard protocols particularly in areas with poor blood flow. Based on our results, (6, 8). a considerably high flushing rate (>30 ml/min) is necessary to avoid charring, which can be a limitation for heart failure HPSD Protocols Comparison patients. The spherical tip also led more frequently to charring, As power is increased and duration reduced shallower lesions are especially in the atrium, possibly related to a larger surface of the generated with minimal change in width. In terms of safety, we tip in contact with blood than with tissue. didn’t find any difference between HPSD protocols in the atrium. Charring mostly appears in our model at the interface between In contrast, in the ventricle, the 90 W/4 s protocol represented the the metallic tip and the rest of the catheter body. We replicate best power/duration balance to ensure safe lesions while the 80 a cooling system based on six distal coplanar pores, with saline W/6 s appeared to be the poorest. The 70 W/8 s protocol offered flowing out perpendicularly to the electrode surface. This is a safe margin only at moderate CF. probably not the most suitable configuration since the electrode- Leshem (6) also identified the 90 W/4 s as the best body interface is not directly cooled down. This advocates for a power/duration combination, allowing them to safely apply CF more efficient cooling system, which could be achieved by either as high as 40 g. Differently to our approach, these authors used adding more pores distributed along the entire metallic tip, or by a highly sensitive and efficient temperature-controlled catheter forcing a directionality in the irrigation flow toward the upper with rapid power downregulation at temperature >65◦ C and a part of the electrode (6, 7). cylindrical tip design with 64-pores, which can explain the wider safety margin with respect to CF. HPSD Lesion Size and Shape Effect of a Second Application In the line of data reported in HPSD in-vitro experiments (6, 7, We observed a substantial increase in lesion size after 24), HPSD lesions are wider than they are deep and, in general, a second identical HPSD application, in opposition to smaller in volume than those of standard protocols. This can be what is expected to occur with standard protocols (28). expected since the total energy delivered by HPSD protocols is According to thermodynamics, heat transfer within the lower than for the standard protocols (e.g., 360 J for a 90 W/4 s vs. tissue continues until thermal equilibrium is reached. This 900 J for a 30 W/30 s). However, the relationship between energy process is estimated to last 45–60 s which is close to the delivery and lesion size is not the same for HPSD and standard end of application in standard protocols. The short duration protocols. Increases in power, combined with shorter duration, of HPSD protocols allows to exploit this thermal latency entail a decrease in depth accompanied by minor changes in period by rapidly adding a second application to increase the width. This can be explained by the distinctive biophysics of heat lesion size. transfer in HPSD protocols, since lesion generation is mainly We analyzed different intervals between applications. The based on resistive heating, while conductive heating has only a optimal interval was a trade-off between efficacy and safety: the marginal role due to the short duration of the application (14). shorter the interval the larger the lesion size gain but, at the same This feature may prove advantageous in the atrium, allowing time, the higher the risk of pops. to treat the arrhythmogenic substrate in its thin wall avoiding Two additional findings should be highlighted. In the atrium, complications in its vicinity. However, it may be insufficient the gain in lesion size was more evident in width than in to reach deep targets as in the ventricle or for mitral annular depth, which would potentially maximize the extension of atrial lines (27). lesions, while preserving neighboring structures. In the ventricle, On the contrary, Leshem et al. (6) reported a better linear the second application achieved lesions comparable in size to continuity and transmurality with HPSD ablations than with standard ones, notably reducing the ablation time. Frontiers in Medical Technology | www.frontiersin.org 10 November 2021 | Volume 3 | Article 747609 Petras et al. Virtual Characterization of HPSD Ablation LIMITATIONS arrhythmogenic substrate preserving the neighboring structures. In the ventricle, a second application can create lesions Our advanced virtual model includes some variables which comparable to the ones obtained with standard protocols in influence the biophysics of RF ablation, making it comparable shorter time. The safety margin of HPSD is narrow but seems to to experimental models. The model assumes constancy of other be a valuable new clinical approach. For best outcomes, it should variables like catheter orientation, that cannot be fully controlled be implemented cautiously and probably using ad-hoc designed in clinical practice and would probably change the characteristics RF systems. of the RF lesion. Due to a lack of data in this respect, we are considering isotropic thermal and electrical conductivities, DATA AVAILABILITY STATEMENT which produce more spherically-shaped lesions. As the cardiac tissue is not isotropic (fibers orientations, tissue composition), All data supporting the findings of this study are available within this modeling choice is likely responsible for the lesion width the article and its Supplementary Material. underestimation we observe with respect to some experimental studies. In our opinion, although deserving further investigation, AUTHOR CONTRIBUTIONS these aspects don’t have a significant impact on the global results of this study since they are common limitations for both standard JG and LG-G conceived the original idea. LG-G, AP, and and HPSD protocols. The main objective of the study was to ML developed the computational model and performed the explore the biophysical characteristics of HPSD in comparison simulations. JG and ZMW assessed on the clinical aspects. AP to standard RF protocols, therefore extrapolation to the clinical and ZMW drafted the manuscript. All authors discussed the setting should be made cautiously. results and contributed to the final manuscript. Our model assumes a power-controlled system, without upper temperature limit. Adding a self-down-regulation of the energy at a given tissue temperature would probably avoid FUNDING some complications observed in our simulations, but would Part of this work was performed when LG-G, AP, and ML prevent a deeper understanding of the behavior of this type of were with the Basque Center for Applied Mathematics (BCAM) energy delivery. in Bilbao, Spain. This research was supported by the Basque The simulated applications were interrupted when blood or Government through the BERC 2018-2021 program, and the tissue temperature reached complication thresholds. Our model Spanish State Research Agency (AEI) through the grant RTI2018- uses a simple six-pores irrigation system prone to charring 093416-B-I00 MULTIQUANT. formation. The question lingers whether by using new catheters with more effective irrigation system charring could have been avoided and if so, whether pops would have occurred had the ACKNOWLEDGMENTS applications continued. According to our simulations, enhanced ML acknowledges LaCaixa 2016 Ph.D. LG-G, AP, and ML irrigation designs would have little impact on the risk of pop or acknowledge the partial support of the State of Upper Austria. lesion size. CONCLUSIONS SUPPLEMENTARY MATERIAL High-power short-duration ablation generates shallower and The Supplementary Material for this article can be found smaller lesions than those of standard protocols. In the online at: https://www.frontiersin.org/articles/10.3389/fmedt. atrium, these characteristics can be exploited to reach the 2021.747609/full#supplementary-material REFERENCES 5. Bourier F, Duchateau J, Vlachos K, Lam A, Martin CA, Takigawa M, et al. High-power short-duration versus standard radiofrequency ablation: 1. Wittkampf FHM, Nakagawa H RF catheter ablation: lessons on lesions. Pacing insights on lesion metrics. J Cardiovasc Electrophysiol. (2018) 29:1570–5. Clin Electrophysiol. (2006) 29:1285–97. doi: 10.1111/j.1540-8159.2006.00533.x doi: 10.1111/jce.13724 2. Kuck K-H, Brugada J, Fürnkranz A, Metzner A, Ouyang F, Chun KRJ, et al. 6. Leshem E, Zilberman I, Tschabrunn CM, Barkagan M, Contreras- Cryoballoon or radiofrequency ablation for paroxysmal atrial fibrillation. N Valdes FM, Govari A, et al. High-power and short-duration ablation Engl J Med. (2016) 374:2235–45. doi: 10.1056/NEJMoa1602014 for pulmonary vein isolation. JACC Clin Electrophysiol. (2018) 4:467–79. 3. Cronin EM, Bogun FM, Maury P, Peichl P, Chen M, Namboodiri doi: 10.1016/j.jacep.2017.11.018 N, et al. HRS/EHRA/APHRS/LAHRS expert consensus statement on 7. Rozen G, Ptaszek LM, Zilberman I, Douglas V, Heist EK, Beeckler C, et al. catheter ablation of ventricular arrhythmias. Europace. (2019) 21:1143–4. Safety and efficacy of delivering high-power short-duration radiofrequency doi: 10.1093/europace/euz202 ablation lesions utilizing a novel temperature sensing technology. Eurospace. 4. Kewcharoen J, Techorueangwiwat C, Kanitsoraphan C, Leesutipornchai (2018) 20:f444–50. doi: 10.1093/europace/euy031 T, Akoum N, Bunch TJ, et al. High-power short duration and 8. Barkagan M, Leshem E, Shapira-Daniels A, Sroubek J, Buxton AE, Saffitz low-power long duration in atrial fibrillation ablation: a meta- JE, et al. Histopathological characterization of radiofrequency ablation analysis. J Cardiovasc Electrophysiol. (2021) 32:71–82. doi: 10.1111/jce. in ventricular scar tissue. JACC Clin Electrophysiol. (2019) 5:920–31. 14806 doi: 10.1016/j.jacep.2019.05.011 Frontiers in Medical Technology | www.frontiersin.org 11 November 2021 | Volume 3 | Article 747609 Petras et al. Virtual Characterization of HPSD Ablation 9. Reddy VY, Grimaldi M, De Potter T, Vijgen JM, Bulava A, Duytschaever 22. Tyler TD. Doppler ultrasound in cardiology, By L. Hatle and B. Angelsen, MF, et al. Pulmonary vein isolation with very high power, short duration, Philadelphia, Lea and Febiger, (1982). 238 pages. J Clin Ultrasound. (1984). temperature-controlled lesions. JACC Clin Electrophysiol. (2019) 5:778–86. 12:244–244. doi: 10.1002/jcu.1870120418 doi: 10.1016/j.jacep.2019.04.009 23. Park C-I, Lehrmann H, Keyl C, Weber R, Schiebeling J, Allgeier J, et al. 10. Wielandts J-Y, Kyriakopoulou M, Almorad A, Hilfiker G, Strisciuglio Mechanisms of pulmonary vein reconnection after radiofrequency ablation T, Phlips T, et al. Prospective randomized evaluation of high power of atrial fibrillation: the deterministic role of contact force and interlesion during CLOSE-guided pulmonary vein isolation: the POWER-AF study. distance. J Cardiovasc Electrophysiol. (2014) 25:701–8. doi: 10.1111/jce.12396 Circ Arrhythm Electrophysiol. (2021). 14:e009112. doi: 10.1161/CIRCEP.120. 24. Kotadia ID, Williams SE, O’Neill M. High-power, short-duration 009112 radiofrequency ablation for the treatment of AF. Arrhythmia Electrophysiol 11. Petras A, Leoni M, Guerra JM, Jansson J, Gerardo-Giorda L. A computational Rev. (2020) 8:265–72. doi: 10.15420/aer.2019.09 model of open-irrigated radiofrequency catheter ablation accounting for 25. Petras A, Leoni M, Guerra JM, Jansson J, Gerardo-Giorda L. Tissue mechanical properties of the cardiac tissue. Int J Numer Methods Biomed Eng. drives lesion: computational evidence of interspecies variability in cardiac (2019). 35:e3232. doi: 10.1002/cnm.3232 radiofrequency ablation. In: Functional Imaging and Modeling of the Heart, 12. Guerra JM, Jorge E, Raga S, Gálvez-Montón C, Alonso-Martín C, FIMH, Lecture Notes in Computer Science, Vol. 11504. Bordeaux (2019) p. Rodríguez-Font E, et al. Effects of open-irrigated radiofrequency 139–46. doi: 10.1007/978-3-030-21949-9_16 ablation catheter design on lesion formation and complications: in vitro 26. Everett TH IV, Lee KW, Wilson EE, Guerra JM, Varosy PD, Olgin JE. Safety comparison of 6 different devices: in vitro comparison of open-irrigated profiles and lesion size of different radiofrequency ablation technologies: a catheters. J Cardiovasc Electrophysiol. (2013) 24:1157–62. doi: 10.1111/jce. comparison of large tip, open and closed irrigation catheters. J Cardiovasc 12175 Electrophysiol. (2009) 20:325–35. doi: 10.1111/j.1540-8167.2008.01305.x 13. Hasgall P, Di Gennaro F, Neufeld E, Lloyd B, Gosselin M, Payne D, et al. IT’IS 27. Yavin HD, Leshem E, Shapira-Daniels A, Sroubek J, Barkagan M, Database for Thermal and Electromagnetic Parameters of Biological Tissues. Haffajee CI, et al. Impact of high-power short-duration radiofrequency Available online at: itis.swiss/database (accessed January 22, 2019). ablation on long-term lesion durability for atrial fibrillation ablation. 14. Haines DE, Watson DD. Tissue heating during radiofrequency catheter JACC Clin Electrophysiol. (2020) 6:973–85. doi: 10.1016/j.jacep.2020. ablation: a thermodynamic model and observations in isolated perfused and 04.023 superfused canine right ventricular free wall. Pacing Clin Electrophysiol. (1989) 28. Wittkampf FH, Nakagawa H, Yamanashi WS, Imai S, Jackman WM. 12:962–76. doi: 10.1111/j.1540-8159.1989.tb05034.x Thermal latency in radiofrequency ablation. Circulation. (1996) 93:1083–6. 15. Rossmanna C, Haemmerich D. Review of temperature dependence of doi: 10.1161/01.CIR.93.6.1083 thermal properties, dielectric properties, and perfusion of biological tissues at hyperthermic and ablation temperatures. Crit Rev Biomed Eng. (2014) Conflict of Interest: JG served as consultant for Biosense Webster, Boston 42:467–92. doi: 10.1615/CritRevBiomedEng.2015012486 Scientific and Abbott, received speaker fees from Boston Scientific and Abbott, 16. Abraham JP, Sparrow EM, A. thermal-ablation bioheat model including and received a research grant from Abbott. liquid-to-vapor phase change, pressure- and necrosis-dependent perfusion, and moisture-dependent properties. Int J Heat Mass Transf. (2007) 50:2537– The remaining authors declare that the research was conducted in the absence of 44. doi: 10.1016/j.ijheatmasstransfer.2006.11.045 any commercial or financial relationships that could be construed as a potential 17. Haines DE. Chapter 1 - Biophysics of radiofrequency lesion formation. conflict of interest. In: Huang SKS, Wood MA, Scheinman MM, editors. Catheter Ablation of Cardiac Arrhythmias. Philadelphia, PA: W.B. Saunders (2006). p. 3–20. Publisher’s Note: All claims expressed in this article are solely those of the authors doi: 10.1016/B978-1-4160-0312-0.50007-4 and do not necessarily represent those of their affiliated organizations, or those of 18. Haines D. The biophysics and pathophysiology of lesion formation during the publisher, the editors and the reviewers. Any product that may be evaluated in radiofrequency catheter ablation. Card Electrophysiol Cell Bedside. (2004). this article, or claim that may be made by its manufacturer, is not guaranteed or 1018–27. doi: 10.1016/B0-7216-0323-8/50114-7 19. Ayachit U. The ParaView Guide: Updated for ParaView Version 4.3. Full Color endorsed by the publisher. Version. Avila L, editor. Los Alamos: Kitware (2015). p. 261. 20. Thompson N, Lustgarten D, Mason B, Mueller E, Calame J, Bell S, et al. The Copyright © 2021 Petras, Moreno Weidmann, Leoni, Gerardo-Giorda and Guerra. relationship between surface temperature, tissue temperature, microbubble This is an open-access article distributed under the terms of the Creative Commons formation, and steam pops. Pacing Clin Electrophysiol. (2009) 32:833–41. Attribution License (CC BY). The use, distribution or reproduction in other forums doi: 10.1111/j.1540-8159.2009.02397.x is permitted, provided the original author(s) and the copyright owner(s) are credited 21. Cabrera JA, Sánchez-Quintana D. Cardiac anatomy: what the and that the original publication in this journal is cited, in accordance with accepted electrophysiologist needs to know. Heart Br Card Soc. (2013) 99:417–31. academic practice. No use, distribution or reproduction is permitted which does not doi: 10.1136/heartjnl-2011-301154 comply with these terms. Frontiers in Medical Technology | www.frontiersin.org 12 November 2021 | Volume 3 | Article 747609