Catheter ablation for atrial fibrillation (AF) has been performed for more than two decades and represents an important treatment milestone for improved symptom and rhythm control. Thermal ablation has been adopted worldwide since the seminal study by Haïssaguerre and colleagues [1] in 1998, which showed that the pulmonary veins (PVs) are a significant trigger of AF. Sham-controlled trial data revealed a substantial reduction in AF burden, improved quality of life, and symptoms [2]. At present, catheter ablation has a class 1A recommendation to improve symptoms in patients with symptomatic AF in whom anti-arrhythmic drugs (AADs) have failed or not been tolerated [3]. Similarly, patients with AF and heart failure with reduced ejection fraction also have a class 1A indication for catheter ablation, given superiority over rhythm control with drug therapy for all-cause death and heart failure hospitalization [3]. Thermal catheter ablation has been undertaken using either radiofrequency (heat) or cryoablation (freezing) with similar outcomes [4]. Nevertheless, these ablative techniques are limited by a lack of selectivity, as these techniques rely on thermal energy to induce cell injury and necrosis. This may lead to PV stenosis and damage to surrounding anatomical structures, including the phrenic nerve and the esophagus. Atrioesophageal fistula, while rare, remains a feared complication, and presents a mortality up to 55% and a lack of definitive preventive strategies [5, 6, 7].

The recent advent of pulsed field ablation (PFA) has changed the therapeutic landscape for catheter ablation for AF. PFA employs cell electroporation and shows greater selectivity for myocardial cells, causes less thermal injury when delivered appropriately, and, thus, mostly spares surrounding extracardiac structures. This differs substantially from radiofrequency or cryoablation, as the anti-arrhythmic effects are achieved not by thermal energy but by electroporation. PFA is also associated with reduced procedural and left atrial catheter dwell times and improved workflow, which facilitates same-day discharge. Growing clinical trial and post-approval registry data have shown that PFA is safe and effective compared to traditional thermal ablation techniques. This has led to increased uptake of PFA worldwide, with more than 500,000 cases performed globally since the first commercial case in 2021.

In this narrative review, we provide an overview of the biophysics of PFA, catheter design, appraisal of clinical trial and registry data, potential specific complications, procedural workflow, and future direction in performing PFA to treat AF (Table 1). Several other narrative reviews have been published in recent years on the use of PFA for cardiac arrhythmias [8, 9]. However, this review focuses specifically on PFA for AF, where the most evidence is available, with a predominant emphasis on clinical data and outcomes (rather than preclinical and benchside data) for clinical cardiologists, and highlights future directions in this field.

PFA relies on cell electroporation to create ablation lesions. An electrical field is generated by the delivery of ultra-short, high-voltage impulses that invert phospholipids at the cell membrane, allowing the passage of ions and small molecules into the cell [10]. Hydrophilic pores form on the cell membrane surface through lipid oxidation, in concert with reactive oxygen species generated by the electric field, disrupting cell proteins and destabilizing the cell cytoskeleton. Other mechanisms of cell injury include Ca²⁺ influx, mitochondrial damage, and adenosine triphosphate (ATP) depletion.

PFA is sometimes incorrectly described as a non-thermal ablation technique. However, electroporation delivered at sufficiently high voltage over a certain pulse duration can still trigger tissue heating. Tissue selectivity applies up to a given threshold of voltage delivery; beyond that, other structures may still be affected. Adjustable parameters of energy delivery include voltage, pulse duration, frequency, biphasic or monophasic energy delivery, and unipolar or bipolar configuration. However, cardiomyocytes may be more susceptible to electroporation due to certain intrinsic cell characteristics, such as relatively greater ion channel expression, which can be denatured by electroporation, and a larger cell radius, which lowers the associated electric field threshold relative to smooth muscle cells (e.g., esophagus) [11]. Furthermore, PFA has a central zone of irreversible electroporation in proximity to the catheter electrode and a surrounding zone of reversible electroporation, where initial cell membrane disruption can recover over time [12]. This represents a potential limitation of the technology, as cells that are only reversibly electroporated may appear electrically silent (acute loss of local intracardiac electrograms) but may still recover over time. Indeed, PVs have been reconnected in up to 55% of patients undergoing clinically indicated repeat procedures [13]. However, this is likely overestimated compared to overall durability when including patients without symptomatic recurrences.

An ever-expanding number of PFA systems are becoming commercially available worldwide. This is a competitive landscape, with the industry driving constant design improvements. However, rather than providing an overview of all commercially available devices or devices currently in clinical trials, understanding the clinical indication for PFA use and the subsequent relevant, tailored design best suited to the purpose is perhaps more important.

In the context of AF, performing PV isolation as the standard de novo ablation strategy is most efficiently achieved through a degree of circumferential design. This enables efficient workflow and potentially mitigates the risk of embolic stroke. Various PFA systems deliver trains of bipolar or biphasic stimuli with an electric field strength of 1.5–2.0 kV over fixed pulse durations. Design differences include the number and shape of electrodes, the relevant electrode spacing, catheter diameter, steerability, and the incorporation of an over-the-wire design. Other factors include integration of tissue-contact sensing and three-dimensional (3D) electroanatomic mapping to guide overlapping lesion delivery, activation, and voltage mapping. While PV isolation remains the standard lesion set in paroxysmal AF, posterior wall (PW) isolation with these catheters is technically feasible.

Focal point-by-point catheters are also currently commercially available or in clinical trials. The ability to deliver focal lesions using PFA will change the treatment landscape for other ablation areas, including the mitral isthmus, cavotricuspid isthmus, and ventricle.

Growing clinical trial data generally support the use of PFA for AF ablation. Initial first-in-human, non-randomized, industry-sponsored studies (IMPULSE and PEFCAT) enrolled patients with paroxysmal, drug-resistant AF to PFA with Farapulse™ (Boston Scientific, USA) [14]. In 81 patients, 87.4% were free of arrhythmias as measured by 24-hour Holter monitoring at 12 months of follow-up. Farapulse™ has also been evaluated in 339 patients with persistent AF who underwent PV isolation and PW ablation in the ADVANTAGE AF single-arm study, achieving a 1-year freedom from arrhythmia of 63.5% [15]. Four patients developed pulmonary edema, likely related to the volume of intravenous fluids administered peri-procedurally to reduce the risk of hemolysis. The PULSED AF Pivotal trial was a non-randomized, single-arm, industry-sponsored trial that evaluated a different circumferential catheter (PulseSelect™, Medtronic, USA) [16]. In 300 patients with paroxysmal and persistent AF, freedom from a composite of acute procedural failure, arrhythmia recurrence, or AAD escalation was 66% and 55%, respectively, at 1 year. Similar findings have been reported with the Varipulse™ (Johnson and Johnson Medtech, USA) circumferential PFA system, with a 1-year freedom-from-arrhythmia rate of 78.9% in 226 patients with drug-refractory paroxysmal AF [17].

Randomized controlled trials have been conducted comparing PFA with conventional thermal ablation (Table 2, Ref. [18, 19, 20]). The ADVENT Trial was an industry-sponsored, non-inferiority trial that randomized patients with drug-refractory paroxysmal AF to either PFA with Farapulse™ versus conventional radiofrequency or cryoballoon catheter ablation [18]. Arrhythmia detection was performed using 72-hour Holter monitoring at 6 and 12 months and weekly trans-telephonic electrocardiographic recordings after a 3-month blanking period. At 1-year follow-up, there was no difference between groups for the primary composite efficacy endpoint of procedural failure, atrial tachyarrhythmia recurrence, AAD use, cardioversion, or repeat ablation. Similarly, the SINGLE SHOT CHAMPION trial was a non-inferiority trial conducted in two centers in Switzerland that randomized patients with paroxysmal AF to PFA (Farapulse™) or cryoballoon ablation (Arctic Front™, Medtronic) [19]. Importantly, all patients had a continuous cardiac monitor implanted, and the trial was investigator-initiated, although the trial received an unrestricted research grant from industry. At 1-year follow-up with a 3-month blanking period, PFA was shown to be superior to cryoballoon ablation for their primary endpoint of freedom from atrial tachyarrhythmia recurrence $\ge$30 seconds (63% vs. 49% at 1-year, respectively; p 0.001 for non-inferiority and p = 0.046 for superiority).

In patients with persistent AF, the SPHERE Per-AF trial evaluated a dual-energy Sphere-9™ catheter with electroanatomic mapping (Affera™, Medtronic) compared with conventional radiofrequency (RF) ablation [20]. In this industry-sponsored, non-inferiority trial with 420 patients, composite freedom from procedural failure and repeat ablation, arrhythmia recurrence (evaluated by 24-hour Holter monitoring), drug initiation or escalation, or cardioversion after a 3-month blanking period was 73.8% and 65.8%, respectively, at 1 year.

While these are well-designed contemporary randomized trials, certain limitations should be recognized. First, these trials possessed relatively small sample sizes, given that more than 500,000 cases have now been performed worldwide using the Pentaspline catheter. Second, these studies are at least, in part, industry-sponsored. Third, a three-month blanking period is likely excessive for PFA technology, given the associated reduced proinflammatory profile compared with conventional thermal ablation. There is also an established association between early arrhythmic recurrences predicting late recurrences [21, 22, 23]. Guidelines now suggest an eight-week blanking period [24]. Fourth, there is heterogeneity in the endpoint, be it arrhythmia recurrence alone or the inclusion of repeat ablation, cardioversion, or drug initiation. Similarly, there is variability in monitoring arrhythmic recurrence, including Holter monitoring, trans-telephonic electrocardiographic monitoring, or an implantable loop recorder. Fifth, these trials have a one-year follow-up. It is recognized that most AF recurrences tend to be frontloaded, potentially arguing for a relatively short duration of follow-up for arrhythmic recurrences [25]. However, longer-term safety outcomes are still needed to assess potential late adverse sequelae, such as coronary artery stenosis [26].

To monitor complications, large post-approval surveys and registry data on the Farapulse™ catheter have reported major adverse event rates of less than 2% and minor adverse event rates of approximately 3–4% [27, 28, 29, 30]. There have been no reported cases of atrioesophageal fistula or symptomatic PV stenosis. Symptomatic phrenic nerve injury has been transient to date. While PFA has been considered a relatively safe ablation modality due to the associated greater tissue selectivity, this technique is not without risk, as PFA is associated with certain potential, if not unique, complications.

Pericardial tamponade has an estimated risk of up to 1.14% across registry data, which places pericardial tamponade as the most common serious adverse complication, occurring more frequently than even major vascular access complications [29]. The risk of pericardial tamponade has been reported to be higher compared with thermal ablation in a meta-analysis of randomized and non-randomized studies (1% vs. 0.2%; odds ratio (OR) 2.98, 95% confidence interval (CI) 1.27–7.00) [31]. The reasons are likely multifactorial, including patient factors, catheter design, operator experience, and therapeutic anticoagulation. Meanwhile, ongoing improvement in catheter system design will likely reduce the risk of inadvertent cardiac injury.

Stroke remains one of the most feared complications related to AF ablation, and this risk persists with PFA. The risk of stroke may depend on the specific PFA catheter and operator experience. Stroke risk does not appear elevated beyond that of conventional thermal ablation, based on post-approval registry and pooled clinical trial data for the Farapulse™ catheter [32]. Nationwide German data reported a stroke rate of 0.2% with thermal ablation and 0.1% with PFA [33]. Conversely, the Varipulse™ catheter was temporarily paused in the United States due to concerns regarding stroke risk. Along the spectrum of stroke, silent cerebral emboli (SCE) have been a concern with conventional thermal ablation, and this risk persists with PFA [34]. Clinical prospective data have reported an SCE risk between 9% and 12% [16, 17]. However, a small study reported SCE in 6 of 7 patients undergoing PFA with Varipulse™, with a median of 13 lesions, $\ge$10 mm lesion size in three patients, and many lesions being multi-territory [35]. This was higher than that reported with PulseSelect, which reported SCE in 2 out of 9 patients (p = 0.04). The mechanism is potentially multifactorial, relating to air entrainment and gas bubble formation during lesion application. Additionally, given the faster procedural workflow for PFA, the peri-procedural activated clotting time may not peak sufficiently post-heparin administration to enable therapeutic anticoagulation for a substantial portion of the procedure. The long-term consequences of SCE on neurocognitive function in PFA remain unknown, and this signal must be carefully monitored.

Coronary artery spasm has been seen with PFA, particularly when used for currently off-label indications such as the cavotricuspid isthmus line in atrial flutter or the posterolateral mitral isthmus line. This phenomenon, rarely observed with current thermal ablation techniques, relates to the proximity of the electric field to the right coronary artery and the left circumflex artery, respectively. The underlying mechanism of vasospasm is thought to be an injection of current via PFA, as well as a calcium imbalance induced by electroporation. Vasospasm tends to last longer than the stimulus and may persist even after calcium levels normalize [36]. In a small study of 26 patients undergoing mitral isthmus line for AF with either PFA or radiofrequency and with concurrent coronary angiography, coronary spasm was observed in 7 out of 17 (41%) patients undergoing PFA [37]. Most cases were subclinical, and two patients received intracoronary nitroglycerin, with spasm resolution occurring within 5 to 25 minutes. However, the subclinical effects could persist up to 3 months post-procedure based on findings from optical coherence tomography data, including reduced arterial luminal area and increased vascular wall area [26]. At present, high-dose (3 mg) intravenous glyceryl trinitrate has often been administered prophylactically by centers that perform off-label ablation near coronary arteries, with reasonable efficacy, although with an expected decrease in blood pressure [38]. Marked troponin elevation is also seen with PFA, even when applied to the PVs and PW, compared with RF. Likewise, this is related to the number of PFA deliveries [39]. The implication of significant troponin elevation in the absence of symptoms or electrocardiogram (ECG) changes is uncertain.

Intravascular hemolysis is a unique phenomenon associated with PFA. Unintentional electroporation of red blood cells during PFA delivery results in hemoglobin release and depends on the strength of the electric field. Biomarkers of hemolysis, including free hemoglobin, haptoglobin, bilirubin, and lactate dehydrogenase, are affected 24 hours post-procedure [40]. While this has not led to meaningful anemia, it can be associated with hemoglobinuria in more than one-third of cases and subsequent acute kidney impairment. There is a strong correlation between the number of PFA deliveries and the risk of hemolysis. Thus, operators are increasingly wary of the number of applications performed in a single procedure. Administration of intravenous fluids has been adopted to reduce the risk of acute kidney injury, particularly in patients with pre-existing renal impairment [41].

PFA has significantly improved the procedural workflow and efficiency of AF ablation. Procedural and left atrial catheter dwell time are reduced, which may mitigate the risk of complications and increase procedural volume. Notably, more than 250 PFA procedures are performed annually at our tertiary center. All patients with AF undergoing their first ablation procedure receive PFA. The operator learning curve is relatively quick, with about 20 cases per operator for single-shot technology; however, improved pulmonary vein isolation (PVI) durability is seen after about 60 cases [42]. Despite increased throughput, financial considerations must be considered, given the higher equipment costs currently associated with PFA [43].

The protocol below is a summarized outline of the clinical experience of our center, while recognizing that other centers may have different practices regarding peri-procedural imaging (e.g., transesophageal guidance), anesthesia and ablation techniques, number of PFA applications, anticoagulation, and hospital discharge protocols.

Patients begin fasting from midnight, and therapeutic anticoagulation is typically uninterrupted. Our ablation procedure is performed under general anesthesia, given the availability of anesthetic support, patient comfort, and the reduced risk of patient movement that could alter electroanatomic mapping, where available. However, many centers perform PFA under deep sedation with comparable safety, faster procedural and laboratory occupancy times, higher patient satisfaction, and no difference in arrhythmic recurrence rates compared with general anesthesia [44, 45, 46, 47].

A cardiac anesthetist routinely performs a transesophageal echocardiogram (TOE) to evaluate the left atrial appendage thrombus, guide transseptal puncture, assess PV anatomy, provide a focused assessment of left atrial size and left ventricular function, and assess for pericardial effusion. We do not routinely perform preprocedural computed tomography (CT); instead, anatomical assessment is guided by TOE.

Two femoral venous access sites are obtained under ultrasound guidance, including one for a coronary sinus decapolar diagnostic catheter for backup pacing and to assist with mapping. Intravenous heparin is administered after femoral access, aiming for an activated clotting time (ACT) of 350 seconds; 10,000 to 15,000 units of heparin is administered upfront to achieve the target ACT before left atrial access is rapidly obtained. Trans-septal puncture is performed under fluoroscopic and TOE guidance with either a standard trans-septal sheath with NRG™ RF needle and over-the-wire exchange or with VersaCross Connect™ (Boston Scientific) when using the Farapulse™ catheter, which eliminates the need for sheath exchange. Atropine is administered due to the potential for a vasovagal response following PFA delivery; we have not routinely observed systemic anticholinergic side effects, such as urinary retention. However, anticholinergic side effects have been reported in prospective observational studies, and glycopyrrolate may be more efficacious (fewer vagal responses, asystole, and the need for temporary pacing) and safer (fewer drug-related events) than atropine in this context [48].

The subsequent procedural workflow is dependent on the PFA system used. Where concurrent electroanatomic mapping is available, the left atrium and PV anatomy can be mapped before ablation, or ablation can be performed after each PV is mapped. We typically commence ablation with the left-sided PVs, given our catheter is usually already on that side immediately after trans-septal puncture (Fig. 1). A post-ablation map is variably performed to assess PV isolation and evaluate gaps (Fig. 2). After catheter withdrawal from the left atrium, the femoral vein is closed with either a figure-of-eight suture or a percutaneous suture closure device such as Perclose ProStyle™ (Abbott Vascular, USA). A recent randomized trial demonstrated shorter time to ambulation and fewer minor vascular access complications with a percutaneous suture closure device compared with standard figure-of-eight suture [49]. However, routine use of a closure device is limited by financial constraints. Patients remain in bed in a horizontal position for 2 to 4 hours, depending on the groin closure technique. Our center routinely discharges patients the same day unless there is a procedural complication.

Fig. 1.

Pulsed field ablation with Farapulse™. Farapulse™ catheter using the Faradrive™ (Boston Scientific) 13 French deflectable sheath with an over-the-wire technique to apply PGA to the left pulmonary vein. An implantable loop recorder is also applied in situ.

Fig. 2.

Pulsed field ablation with Faraview™ electroanatomic mapping. Bipolar voltage map of the left atrium and pulmonary veins in posterior-anterior (PA) (left side) and Left anterior oblique (LAO) view (middle panel) with local electrogram signals displayed on the right-side panel. The colors represent varying levels of local bipolar voltages: purple indicates areas of high bipolar voltage signals, while red indicates areas of low bipolar voltage signals, correlating with prior ablation.

While the published literature regarding PFA for AF is rapidly expanding, many clinical questions remain unanswered. Thus, future research priorities should include (1) evolving catheter design, (2) ongoing monitoring for rare and late complications, (3) lesion optimization and appropriate ablation strategy, particularly for persistent AF, and (4) safety and efficacy of PFA in specific patient groups (Table 1).

PFA represents a new horizon in the treatment of AF. Procedural efficiency improves, and efficacy is comparable to that of conventional thermal ablation. Early safety data appear acceptable, although specific, uncommon complications unique to PFA require further attention. Given this novel technology, the threshold for offering patients an AF ablation has changed, which has important implications for tackling the burden of AF in our community and improving clinical outcomes for our patients.

FJH, HCH and EK designed the review. NN and AJB contributed the conception of the manuscript. FJH designed the tables, figures and analyzed the data. FJH drafted the manuscript. All authors (FJH, HCH, NN, AJB, EK) contributed to critical revision of the manuscript for important intellectual content. All authors read and approved the final manuscript. All authors have participated sufficiently in the work according to ICMJE guideline criteria.

Not applicable.

Not applicable.

This research received no external funding.

E.K. reports serving on the medical advisory boards for Medtronic, Boston-Scientific, and Biotronik. Other authors do not have relevant disclosures.