High-power short duration ablation for pulmonary vein isolation: simply cranking up the energy?

This editorial refers to ‘Safety and outcome of very high-power short-duration ablation using 70 W for pulmonary vein isolation in patients with paroxysmal atrial fibrillation’, by M. Kottmaier et al., on pages 388--393. Over the last 2 years, pulmonary vein isolation (PVI) using higher power with shorter duration has gained growing interest. Whereas conventional irrigated tip radiofrequency (RF) ablation is performed using 20–40 W applications, significantly higher power (50–90 W) for shorter duration pulses (4–7 s) have been published in experimental and clinical studies.1–4 Since the initial study by Nilsson et al.5 on high output/short-duration ablation, reports on experimental settings and clinical ablation of more than 12 000 patients (mostly from two publications) using high-power short duration (HPSD) have been published with a diverse matrix of parameters (such as delivered energy, duration of RF pulses, mode of ablation generator, temperature targets, and catheter specifications).1–4,6–9 As a general consent high-power atrial ablation is defined by energy applications ≥45 W and short duration usually refers to ablations of 20 s or less, but no clear definition exists. Clinical studies on HPSD using 50 W (5–7 s) or 90 W (4 s) have demonstrated high efficacy and comparable or superior safety in PVI with considerable shorter total RF and procedure durations.1,2 Kottmaier et al.10 add more clinical data on a novel parameter combination of HPSD using 70 W combined with short ablation pulses (restricted to 5 s on the posterior and 7 s on the anterior left atrial wall) in 97 patients with paroxysmal atrial fibrillation. They compared 12-month clinical outcome and complications in the HPSD arm to a historical 30–40 W for 20–30 s ablation cohort of 100 patients. In summary, the HPSD group had significantly shorter RF (∼1/3) and procedure duration (minus 20 min) compared to the conventional group. Patients in the HPSD group had significantly less adenosine-detected dormant PV conduction (13% vs. 55% of PVs) after PVI and significantly fewer atrial arrhythmia recurrences (14.2% vs. 32.3%) during 12-month follow-up. No procedure-related complications including pericardial tamponade, periprocedural thromboembolism, or atrio-oesophageal fistula (AEF) were documented.10 The authors are to be congratulated on the first report on the clinical efficacy of HPSD using 70 W/5–7 s PVI ablation. In congruence with other reports on different HPSD settings (45–50 W/2–15 s up to 90W/4 s), procedure and RF application durations were shorter and outcome data indicate a comparable or superior acute and mid-term efficacy with a superior safety profile.1,2,10 However, in contrast to other studies on HPSD, no contact force (CF) measuring catheter or specific temperature monitoring facilities of the ablation catheter were used in this set of patients thereby preventing the reproducibility by less experienced operators and centres. Whether automated lesion formation tools from different manufacturers such as force-time integral, lesion size index, or ablation index may be helpful for stable lesion generation in HPSD currently remains unclear and these algorithms were not used in the current study. The advantages of HPSD over conventional ablation are determined by the biophysics of lesion generation and lesion geometry. High-power short duration creates a higher thermal gradient and thereby reduces the time needed to achieve a lesion in the targeted depth, thereby leading to a change in resistive and conductive heat transformation. Consequently, lesions are superficially wider but less deep compared to conventional settings. This serves for increased continuity of linear lesion ablation and may help to reduce collateral damage by decreased depth of individual lesions. On the other hand, these advantages come with some limitations as HPSD appears to have a narrow safety and efficacy window: increasing ablation duration may lead to excessive heating with creation of steam-pops, thrombus-formation, and excessive lesion depth whereas too short duration will lead to superficial and non-transmural lesions. As duration of RF pulses is the most important determinant automated control of RF duration may be needed to optimize RF energy applications per lesion.7–9 Kottmaier et al.,10 known as a very experienced ablation centre, have not used CF in this study as the ablation catheter used (Flexability, Abbott) does not allow for monitoring CF. Whether this catheter is ideal for HPSD remains unclear as it compromises CF for a high-irrigation profile and distally located temperature sensing, also relevant for safe high-power ablations. In their study, they have been able to perform effective and safe ablations without advanced monitoring and lesion size prediction capabilities (such as lesion size index, ablation index which essentially integrate contact with power and RF duration) and without automated RF application duration control. To our mind, CF feedback seems essential for RF ablation procedure reproducibility and in further consequence to transfer any applied technique to other, less experienced centres. In addition, as RF duration for high-power ablations is the critical determinant of lesions formation, applying methods to determine the effectiveness of local ablation are recommended over simple time cut-offs. Subtle changes in RF application duration may relevantly affect lesion geometry. The relative contribution of small changes in timing if not using automated algorithms for each individual ablation site remains unknown and specifically inadvertently adding RF duration may increase applied energy (for 0.5 s longer duration up to 10% more energy).6–9 Contact—another critical component of lesion formation—may be more critical to perform contiguous and transmural lesions with conventional ablation but even short episodes of insufficient catheter stability during HPSD may—relatively to the overall duration of a RF pulse—have relevant effects on lesion formation. We should make sure that, if HPSD becomes routine practice in a large variety of centres, automated RF generator programs for application are available to reproduce protocols and create optimal efficacy and safety.2,6–9 It appears clear that identifying the optimal settings for HPSD including RF energy, duration, CF, temperature control over the catheter, power- or temperature control mode, irrigation flow, and others are critical for safe, effective, and reproducible PVI. In a recent experimental study, testing different energy-duration combinations 90 W for 4 s was the best compromise between safety and efficacy.8,9 The QDOT-FAST trial clinically investigated this very HPSD ablation strategy using a dedicated new-design of ablation catheter with distal temperature sensors capable of fast and adequate energy titration response. In contrast to other studies on HPSD, a temperature-controlled mode with a temperature cut-off of 65°C was used which may have thermodynamic advantages.2 Published results indicated high efficacy and safety of this specific protocol. A major limitation of the study by Kottmaier et al.—as well as others on clinical outcome of HPSD—is the lack of information on asymptomatic complications like silent cerebral events/lesions and endoscopically detected thermal oesophageal lesions considered to be an indicator of the risk for rare complications such as thromboembolism/stroke and AEF.1–4,10–12 In a large observational multicentre study by Winkle et al.,1 one AEF (incidence 0.009%) was documented using HPSD ablation (45–50 W) and even though the incidence appears to be lower compared to conventional ablation (by factor 13) critical awareness and rigid study protocols are needed when evaluating modifications of routine ablation parameters. In the QDOT-Fast trial, all patients underwent post-ablation brain magnetic resonance imaging (MRI) to test for silent cerebral lesions and the incidence was documented to be comparable to incidences in conventional ablation studies (12%).2 In an evaluation of post-ablation MRI on HPSD (50 W/5 s), no difference in oesophageal enhancement pattern in 574 patients compared to low-power long-duration ablation patients (n = 113) was identified.4 On the other hand, post-ablation endoscopy oesophageal thermal damage was detected significantly less often in 60 W short-duration ablations.3 These data propose a better safety profile of HPSD but no consistent evaluation of all potential side effects is available. Due to the multitude of differences in parameters used in clinical studies for HPSD, comparison in between protocols is difficult and transferring risk assessments from one to another combination of parameters is inadequate. Preventing relevant complications such as AEF remains most important when striving for higher efficacy of PVI. Monitoring of oesophageal temperature using currently available oesophageal probes with their immanent temperature measurement latency is inappropriate for HPSD. The higher the applied power the less latency of monitoring and safety algorithms is acceptable.13,14 In addition, with HPSD basic principles like reducing application of total energy at the posterior left atrial wall (shorter duration to reduce conductive heat transfer) should be acknowledged and may imply automated control of RF duration and using indices of individual ablation lesion formation. In summary, the study by Kottmaier et al.10 adds valuable information on modifications of HPSD using 70 W applications indicating high efficacy and shorter procedural duration. From all data available today, no escalated complication rates are detected when using a wide variety of HPSD settings but vigilance to asymptomatic indicators of the safety profile is recommended especially as subtle differences in parameters may have relevant effects in short-duration high-power RF applications.1–4,10 The best set of parameters optimizing safety and efficacy may include a range of combinations of energy and duration but today remains unclear and fine-tuning is essential. Most ablation specifications used for HPSD today are based on the capabilities of the used RF generators and are not driven by scientific evidence. As we have learned from studies like the one by Kottmaier et al.,10 there is more to HPSD than just simply cranking up energy. Exact titration of RF application duration to the target energy, using decisive novel catheter and RF generator designs to accommodate for fast reliable temperature and energy control, dedicated clinical studies applying to strict, approved protocols and also consequent evaluation of clinical and subclinical side effects will help to lead HPSD to prime time in everyday practice. Conflict of interest: H.P. has consultancy agreements with Biosense Webster and Abbott. T.D. and P.H. have no conflicts of interest. The opinions expressed in this article are not necessarily those of the Editors of Europace or of the European Society of Cardiology.