Lesion metrics and 12‐month outcomes of very‐high power short duration radiofrequency ablation (90W/4 s) under mild conscious sedation

Pulmonary vein isolation (PVI) is often performed under general anaesthesia (GA) or deep sedation. Anaesthetic availability is limited in many centers, and deep sedation is prohibited in some countries without anaesthetic support. Very high‐power short duration (vHPSD—90W/4 s) PVI using the Q‐Dot catheter is generally well tolerated under mild conscious sedation (MCS) though an understanding of catheter stability and long‐term effectiveness is lacking. We analyzed lesion metrics and 12‐month freedom from atrial arrythmia with this approach.

no procedural related complications, and 12-month freedom from atrial arrhythmia was observed in 78.4%.
Conclusion: vHPSD PVI is feasible under MCS, with encouraging acute and longterm procedural outcomes.This provides a compelling option for centers with limited anaesthetic support.
ablation, atrial fibrillation, conscious sedation, pulmonary vein isolation, very-high power short duration

| INTRODUCTION
Pulmonary vein isolation (PVI) remains the cornerstone in atrial fibrillation ablation. 1High power short duration (HPSD) radiofrequency (RF) ablation is proven to be safe and effective during PVI. 2 This typically involves point-by-point ablation using 50 W power under general anaesthesia (GA) or deep sedation 2 .It is generally believed that GA provides an advantage to the operator in terms of catheter stability due to mechanical ventilation allowing predictable chest movement, without sudden excursions due to pain. 3 Many centers utilize GA as standard for AF ablation, and prior studies suggest this may improve outcomes. 4However, RF technology has advanced over the past decade, and anaesthetic support is not always available, especially in publicly-funded healthcare systems.Deep sedation, using a propofol infusion, is a common alternative to GA, however many countries restrict such approaches to anaesthetic-supported cases only.In a recent global survey of 297 centers, one in three ablations were performed using nurseadministered mild conscious sedation (MCS), as almost 60% of centers reported that anaesthetic presence was mandated to perform deep sedation. 5The ability to perform high quality, well-tolerated, PVI under MCS is therefore compelling for electrophysiologists working in centers where access to anaesthetic support is limited.
The Q-Dot ablation catheter (Biosense Webster) delivers contact force (CF) guided temperature-controlled RF energy with the option to administer very-high power short duration ablation (vHPSD) as a fixed prescription of 90 W for 4 s. 6The safety and efficacy of 50 W and 90 W PVI is comparable, 7 however studies comparing these approaches utilize GA.We have previously shown that patient tolerability and acute procedural outcomes of PVI using vHPSD ablation under nurseadministered MCS is comparable to cryoballoon ablation. 8However, catheter stability and lesion delivery metrics under MCS have not been studied, nor have longer-term outcomes of this approach.
To address this uncertainty, we studied ablation lesion metrics, first pass isolation (FPI) rates and real-world outcomes in a consecutive cohort of patients undergoing first time PVI using vHPSD ablation under MCS at our center (Liverpool Heart and Chest Hospital NHS Foundation Trust, United Kingdom).We compared our findings to a contemporaneous cohort of patients undergoing conventional (50 W) HSPD ablation under GA.We sought to determine the percentage of effective, and potentially ineffective, lesions delivered, as a marker of catheter stability.This may be inferred through impedance drop, with studies suggesting that a drop of ≥10 Ω is ideal, and <5 Ω is usually inadequate. 9Hence, we calculated the percentage of lesions per case with <5 Ω and ≥10 Ω impedance drops, respectively.

|
The "Liverpool Streamlined Approach" to vHPSD PVI (90 W/4 s) under MCS Our approach is summarized in Figure 1.Patients with paroxysmal AF were preferred for this approach, in the hope that they would be in sinus rhythm on the table and avoid procedural delays organizing electrical cardioversion via the on-call anaesthetist.Advance communication warning patients they may experience some pain during the procedure was essential.Patients requesting GA or considering themselves to have a "low pain threshold," or with BMI > 35, or with a history suggestive of sleep apnoea were not considered for this approach.Intravenous opiates (25-50 mcg fentanyl) with or without benzodiazepines (1-2 mg midazolam) were administered by nursing staff trained in management of sedation, and under physician direction, without an anaesthetist in attendance.Supplementary oxygen was delivered via nasal cannula at 2-4 L/min, with continuous monitoring of blood pressure, oxygen saturation, heart rate and ECG.Further top-up doses of fentanyl and midazolam were administered at the start of the first RF application, and to control pain and discomfort during ablation as required.Equipment needed for resuscitation was available in the EP laboratory, with the on-call anaesthetist available in case of emergency airway issues or to facilitate electrical cardioversion following ablation if necessary.
RF PVI under MCS in our center employs a single catheter technique, an approach we have previously described. 10,11After a single transseptal puncture, a steerable sheath (Agilis, Abbott) is utilized for catheter navigation in all cases.Mapping of left atrial anatomy is performed with the ablation catheter only.Bilateral wide antral circumferential ablation (WACA) is performed point-by-point to isolate both pulmonary vein pairs.vHPSD RF is delivered with target inter-lesion distance of 3−4 mm anteriorly and 5−6 mm posteriorly, with successful PVI determined by entrance and exit block.The right-sided WACA lesion set is delivered first, to allow time for potential reconnections to emerge, as these are more frequent on the right side due to nearby epicardial fibers.For the same reason, empirical right-sided carinal lines are often delivered.
The posterior aspect of the left WACA is usually ablated last when patients were most sedated, as this can be more painful due to oesophageal proximity.Oesophageal temperature monitoring is not used.To avoid excessive catheter movement, patients are encouraged to lightly breath-hold at end-inspiration before the delivery of each 4 s lesion, with adequate breathing recovery time between each lesion.Effective patient communication is vital during the procedure for continuous assessment of the level of sedation, 12 as well as to keep the patient reassured, and to avoid map shifts from sudden movements due to unexpected pain.If required, touch-up lesions using 50 W power can be delivered using the same Q-Dot catheter.

| RF ablation in the 50 W/GA cohort
Patients with persistent AF were preferentially selected for HPSD ablation under GA due to easy facilitation of electrical cardioversion with continuous anaesthetic support.Patients were anaesthetized, intubated, and ventilated before procedure start, and an anaesthetist and supporting staff were present during the entire procedure.A steerable sheath (Agilis, Abbott) was placed in the LA following a single transseptal puncture.RF was delivered according to the CLOSE protocol at 50 W in all areas, guided by Ablation Index (typically 400 posteriorly, with caution not to exceed 10 s on the posterior wall; 500−550 anteriorly) with continuous oesophageal temperature monitoring-ablation was discontinued if temperature exceeded 38°C and only recommenced when it had returned to baseline.Automated lesion tagging (VisiTag TM ) was employed with the F I G U R E 1 Liverpool streamlined vHPSD protocol.AF, atrial fibrillation; BMI, body mass index; DCCV, direct current cardioversion; LA, left atrial; PVI, pulmonary vein isolation, RF, radiofrequency; vHPSD, very high power short duration ablation; WACA, wide area circumferential ablation.
following settings: (a) diameter of 3 mm, (b) 3 mm catheter stability for minimum 3 s, (c) minimum 3 g CF for >30% time.A maximum inter-tag distance of 6 mm on posterior wall and 5 mm on anterior wall was maintained to ensure lesion contiguity.
Both arms utilized the nMarq or nGen RF generator (Biosense Webster).The reference electrode was positioned on the patient's lower back.During ablation, CF and circuit impedance were monitored.

| Lesion metrics and outcome measures
The following lesion metrics were exported via the CARTO VisiTag module and calculated for each case: Nurse.If symptoms persist beyond the 3-month blanking period, standard practice is to arrange ambulatory monitoring (up to 2 weeks), to ascertain that symptoms are due to AF recurrence.In some cases, patients utilize their own wearable devices, and traces are emailed and uploaded to our patient records for review.

| Statistical analysis
Continuous variables were reported as mean ± standard deviation or median (25th quartile-75th quartile) and were compared using t-tests or nonparametric equivalents depending upon distribution.Categorical variables were reported as counts and percentages and compared using Fisher's exact test.Two-tailed p < .05were considered statistically significant.CARTO data were extracted using bespoke code written in Python.Statistical analysis was performed in R.

| Patient cohorts and characteristics
Eighty-three patients were identified between March 2021 and December 2022 that met our strict inclusion criteria (51 patients 90 W/MCS; 32 patients 50 W/GA).The low number of GA cases was a result of the restriction in anaesthetic availability during the COVID pandemic coinciding with the start of our study period when the Q-Dot catheter was made available at our center.The patient populations were comparable in terms of age, sex, body mass index, and CHA 2 DS 2 -Vasc score, and were balanced on most comorbidities, the main exceptions being hypertension (31.4% vs. 56.2%;p = .022),persistent AF (15.7% vs. 56.2%;p < .001)and moderate-severe left atrial enlargement (15.7% vs. 34.4%;p = .062).Demographic and clinical differences between cohorts are shown in Tables 1 and 2.
None of the 90 W/MCS cohort required conversion to GA. Bilateral FPI rates were similar (64.7% vs. 71.9%;p = .631)as were individual WACA FPI rates (left 82.4% vs. 87.5%,p = .758;right 74.5% vs. 78.1%,p = .796;90 W/MCS vs. 50 W/GA, respectively).Two 90 W/MCS cases received touch-up lesions with 50 W ablation to achieve PVI.PVI was achieved by the end of every case in both cohorts.There were no issues with 3D map stability owing to ablation related discomfort.

| Clinical outcomes
There were no acute complications in the 90 W/MCS cohort.Specifically, there were no steam pops, perforations, oesophageal injury, or thrombo-embolic complications.A single complication (major femoral vascular injury requiring surgical intervention) occurred in the 50 W/GA group, despite ultrasound-guided access.
No map shifts were recorded in either cohort.
Across median 12-month follow-up, freedom from atrial arrhythmia in the 90 W/MCS cohort was observed in 78.4% of patients, compared with 68.8% in the 50 W/GA group (p = .437),the difference likely being accounted for by the prevalence of persistent AF, hypertension and significant left atrial enlargement in the 50 W/GA group.
Redo ablation was required in 3 (5.9%) of the 90 W/MCS group versus 2 (6.2%) of the 50 W/GA group (p > .999)during 12-month follow-up.Reconnections were found in all redo cases.In the 3 90 W/MCS redo cases, reconnections were all right-sided, whilst the left inferior PV was reconnected in 2 of 3 50 W/GA cases.An example of redo ablation 6 months following 90 W/MCS ablation is presented in Figure 2. Left sided FPI was achieved and remained durably intact at redo.However, right sided FPI was not achieved, with further ablation required at the posterior carina (likely endocardial to the septo-pulmonary bundle) with reconnection at this site found at redo.

| DISCUSSION
The main findings from our study were: 1. Using our "Liverpool streamlined approach", vHPSD ablation (90 W/4 s) can be performed under nurse-administered MCS safely and effectively to achieve adequate rates of 12-month arrhythmia freedom.
T A B L E 1 Demographic and clinical differences between patients undergoing vHPSD versus standard RF ablation.3. Whilst more lesions may be required using 90 W/MCS to achieve FPI, total ablation duration is significantly shorter and procedural times are similar when compared to 50 W/GA.

| Why perform PVI under MCS?
The notion of performing RF PVI without an anaesthetist to support GA or deep sedation is unfamiliar in many parts of the world, but very much a reality for others.The move within the EP community to pulsed field ablation for PVI is exciting, but this currently mandates a Median used for <5 Ω impedance drop due to data being highly right-skewed-that is, most cases had very few.
b CTI, Cavotricuspid isthmus; lesions were performed with standard RF in both arms and CTI lesions were excluded from other calculations, hence ablation duration, impedance and contact characteristics are reported for pulmonary vein isolation only.GA, general anaesthesia; MCS, mild conscious sedation.Total cath lab time refers to the time from the patient entering to exiting the lab.Skin-to-skin time refers to the time from first needle puncture to sheath removal.Set-up time refers to the time from lab entry to needle puncture.Turnaround turn refers to time from sheath removal to the patient leaving the lab.

F I G U R E 2
An example of a 90 W/MCS case who underwent redo ablation.Far left: index procedure showing left-sided vHPSD Tags; first pass isolation was achieved.Center left: The left-sided WACA remained durably intact at redo ablation using the Ensite-X mapping system.Center right: index procedure showing right-sided vHPSD Tags; first-pass isolation was not achieved and required additional posterior carinal ablation and a carinal line.Far right: Redo ablation after 6 months using the Ensite-X system demonstrated reconnection at the right posterior carina, which was successfully re-isolated.
GA or deep sedation.For us, the desire to perform PVI under conscious sedation using vHPSD was a logical consideration during the COVID-19 pandemic, when access to the support of an anaesthetist was limited, as they were redeployed to intensive care.
In the aftermath of the pandemic, ablation under MCS remains attractive as it offers a streamlined approach to ablation without the need for anaesthetic support, providing cost saving benefits in a publicly-funded health service. 13The MCS approach also reduces the emission of anaesthetic gases, which may have adverse environmental effects. 14r results add to our prior work using the Q-Dot catheter, suggesting vHPSD ablation under MCS using our approach is not only safe and generally well-tolerated, 8 but offers similar lesion metrics, catheter stability and long-term clinical outcomes to 50 W ablation under GA.Indeed, our arrhythmia freedom rates were similar to those seen in the Q-FFICIENCY trial, which utilized vHPSD ablation under GA and reported 82.1% 12-month arrhythmia freedom. 15propriate patient selection is essential, as is advance communication, warning patients they may experience some pain during the procedure, to avoid MCS in those more anxious or considered to have a lower pain threshold.Similarly, those with potential airway risk, for example morbidly obese patients or those with sleep apnoea, are unsuitable for this approach.
Notably, total cath lab times were similar between 90 W/MCS and 50 W/GA cases.Whilst it may be assumed that MCS cases would be shorter, as this approach avoids delays waiting for GA induction and recovery, MCS also necessitates intraprocedural delays to allow for analgesia and sedation to take effect.Additionally, more lesions were required in the MCS group (87 vs. 58; p < .001),further prolonging the procedure.

| Improving FPI rates
FPI is recognized as a surrogate intraprocedural marker for robust PVI and fewer arrhythmia recurrences. 16,17The biophysics of lesion formation at 90 W are believed to result from the replacement of reversible conductive heating with increased irreversible resistive heating at the catheter-tissue interface, leading to superior line contiguity in bench studies. 18,19This promising profile unfortunately failed to provide satisfactory FPI rates in the clinical setting with the standard CLOSE approach (5−6 mm inter-lesion distance). 20,21bsequent animal studies demonstrated how a 90 W/4 s lesion delivers around 338 J of energy to the tissues; significantly less than a 50 W/10 s lesion (490 J), with less resistive heating, resulting in smaller lesions in both width and depth. 22We adapted by clustering our vHPSD lesions closer together (3−4 mm inter-lesion distance), on the anterior wall and carina, an approach that was elegantly shared in the "very close" vHPSD protocol. 23In our study, this resulted in similar FPI rates between arms.
Redo ablation was required in 5.9% of the 90 W/MCS group, and reconnections were more frequently noted in the right-sided pulmonary veins.This may reflect insufficient lesion depth with 90 W/4 s to address epicardial connections into the pulmonary veins (e.g., the septo-pulmonary bundle).Two patients in the 90 W arm required additional 50 W lesions to achieve PVI, where there was concern of insufficient lesion depth, and the versatility to employ this hybrid approach is a particular strength of the Q-Dot catheter.

| Catheter stability with 90 W/4 s ablation
Under MCS, catheter stability may be worsened by movement due to pain, and unpredictable respiratory motion.A lesion limited to just 4 s may negate this concern.However, a single second of lost contact represents a quarter of the ablation time.Our approach mandates a steerable sheath to support catheter stability, and we encourage patients to lightly breath-hold before the delivery of each 4 s lesion to avoid excessive catheter movement.In doing so, we observed similar mean CF values and generator impedance drops between 90 W/MCS and 50 W/GA lesions, and we did not record any map shifts due to patient movement.An impedance drop < 5 Ω is thought to represent ineffective lesion delivery, and the median percentage of lesions failing to achieve 5 Ω drop with 90 W/MCS was only 2.5%.
Interestingly, the percentage of lesions failing to meet this target was significantly higher with 50 W/GA (13.7% vs. 2.5%; p < .001).Whilst this might suggest more ineffective lesions occur with 50 W, perhaps from greater catheter instability associated with longer lesions, some alternative explanations are possible.Firstly, this may be explained by cohort disparities in our study groups-that is, delivering ablation within a dilated, scarred atrium (as was more common in the 50 W/ GA group) may affect the ability to achieve effective impedance drops.Alternatively, this may be explained by limitations in our understanding of "impedance drop" as a marker of lesion durability, where prior studies have suggested numerically higher impedance drops with 90 W/4 s as compared to 50 W/12 s ablation, a result of faster time to reach 50°C. 24r study demonstrates that vHPSD (90 W/4 s) under MCS is feasible.Future randomized studies comparing procedural and longterm outcomes of this approach with (a) vHPSD (90 W/4 s) under GA, as well as (b) HPSD (50 W) under MCS would allow better understanding of the influence of anaesthesia and power on outcomes, and aid in determining which approach is more effective.

| LIMITATIONS
This was a retrospective, nonrandomised, single-center study with limited patient numbers and therefore underpowering is possible.However, our study presents unique, hypothesis generating data from a novel ablation technology and future research may build upon this.Arrhythmia recurrence rates may have been underestimated with routine follow-up rhythm monitoring limited to a 12 lead ECG at 4 and 12 months only (outside of symptom triggered assessments).This study should not be relied upon to compare long term outcomes between 90 W/MCS versus 50 W/GA as the patient CALVERT ET AL.
| 1171 populations differed significantly in terms of AF classification, and our study was not powered to detect differences in arrhythmia recurrence rates.
The data in this study also may not translate to other vHPSD systems (e.g., 70 W for 7 s), nor to other ablation catheters.
Additionally, our novel metrics (percentage of lesions with <5 Ω and >10 Ω impedance drops) have not been validated in prospective studies but were included as logical concepts which are easily calculated from data exported from CARTO.
During the course of this study period, we found our 90 W ablation tag location on the 3D map not always representative of the actual ablation location.This was because each tag location was adjusted based on respiratory movement, which requires a full breath cycle to correct-something which would not always occur within a 4 s ablation.An "Enhanced Stability algorithm" has since been developed, which facilitates respiratory compensation during 90 W ablation.This requires prospective testing.

| CONCLUSION
This study was a single-center retrospective cohort analysis of RF PVI performed using the Q-Dot catheter.We studied cases performed between March 2021 (acquisition of the Q-Dot catheter at our center) and December 2022.Only patients undergoing Q-Dot ablation at 90 W under MCS, or 50 W under GA, were included and followed-up for a maximum of 12 months.The study was approved by our local Research & Innovation Committee.Exclusion criteria included redo procedures and cases involving non-PV ablation within the left atrium (cavotricuspid isthmus [CTI] ablation was allowed but CTI lesions were excluded from analyzed lesion statistics).Patients were stratified by ablation type (90 W/ MCS vs. 50 W/GA cohorts).Demographic and clinical data were manually extracted from our electronic patient records.Lesion data for each patient were exported from the CARTO platform VisiTag module.This system records lesion-specific parameters during ablation, such as lesion duration, CF and impedance changes during delivery.

1 .
Mean impedance drop 2. Mean CF 3. Total number of lesions delivered 4. Percentage of lesions with <5 Ω impedance drop 5. Percentage of lesions with >10 Ω impedance drop Lesions with missing data (due to failure to record on CARTO) were excluded from analysis.This applied to no lesions in the 90 W/MCS arm, and median 2 (IQR 1−3.75) lesions per patient in the 50 W/GA arm.Clinical outcomes of interest were: 1. FPI rates 2. Procedural time-recorded as the time for the patient to enter and exit the catheter lab 3. Acute procedural complications 4.After a 3 month blanking period, documented arrhythmia recurrence up to 12 months postablation 5. Subsequent redo ablations Class I/III antiarrhythmics were stopped after a 3 month blanking period.To determine arrhythmia recurrence during follow-up, we reviewed all electronic patient record encounters.This included scheduled specialist arrythmia clinic reports and 12 lead ECGs both 4 and 12 months postablation, and any unscheduled symptom triggered telephone encounters with our Heart Rhythm Specialist Ablation delivery time was significantly shorter in the 90 W/MCS group (352.9 s vs. 885.6s; p < .001),however overall procedural times were similar (149.3 vs. 149.1 min; p = .981),perhaps because significantly more lesions were delivered per patient in the 90 W/ MCS group (median 87 vs. 58; p < .001)with narrower inter-lesion distance targets.Mean skin-to-skin time was slightly longer in the 90 W/MCS group but did not meet statistical significance (114.9 vs. 105.1 min; p = .256),likely due to the requirement to wait between top-up sedation/analgesic doses during MCS cases.Median set-up time (time from the patient entering the lab until first needle puncture) was significantly shorter in the 90 W/MCS group (17 vs. 25 min; p < .001).Turnaround time (time from sheath removal until the patient leaves the lab) was similar between groups (mean 16.7 vs. 15.9 min; p = .617).In the 90 W/MCS arm, the mean Midazolam dose was 1.3 mg ± 1.1 mg (range 0−4 mg) and the mean Fentanyl dose was 156 mcg ± 49.7 mcg (range 50−300 mcg).
PVI using vHPSD ablation (90 W/4 s) adopting our Liverpool streamlined approach under nurse-administered MCS can be performed both safely and effectively.Utilizing 3−4 mm anterior wall inter-lesion spacing, more lesions are required to achieve FPI compared to traditional 50 W ablation, but as overall ablation times are much shorter, procedural times are similar.Individual lesions achieve adequate CF, stability and impedance drop, with encouraging 12-month arrhythmia freedom.These findings are compelling for centers where access to anaesthetic support to offer deep sedation or GA is limited.
Procedural Differences between patients undergoing vHPSD versus standard RF ablation.
Note: Bold values indicate statistically significant p values.