Irreversible Electroporation Ablation for Atrial Fibrillation: Status and Challenges

Abstract

Pulsed electric field (PEF) is a novel energy source by which high-voltage electric pulses are used to create irreversible electroporation. PEF is non-thermal and highly tissue-dependent in which specific targeting of the atrial myocardium is achieved and sparing of adjacent tissues is feasible, theoretically increasing the safety of the procedure, which could potentially break the trade-off between effective lesions and collateral damage and substantially improve risk-benefit ratios in atrial fibrillation (AF) ablation. Although recently published trials have shown a clear effect of AF ablation, large-scale clinical trials are lacking. Current clinical evidence has demonstrated significant efficacy in achieving durable pulmonary vein isolation without ablation-related adverse events. However, their putative benefits regarding efficacy, efficiency, and safety remain to be proven in randomized controlled trials.

Atrial fibrillation (AF) and its treatment using radiofrequency (RF) and cryoballoon (CB) ablation

AF is the most common arrythmia and is characterized by rapid and irregular beating of the atrial chambers of the heart. This disrupted rhythm occurs because of the irregular and unpredictable conduction of disordered impulses originating in the atria, which cause uncoordinated contractions between the atria and ventricles. The ineffectual contractions result in lower blood output from the atria into the ventricles, and increases the vulnerability to thrombus formation owing to blood pooling in the atria. This thrombus can then travel to the brain and block blood flow to a part of the brain, resulting in a stroke. Indeed, AF is a major risk factor for ischemic stroke and is associated with a 5-fold higher risk of stroke.^[1] It is estimated that AF affects more than 3 million people in the US,^[2] and more than 33 million people worldwide.^[3] The prevalence of AF increases to more than 1 in 10 in the elderly, and the overall prevalence will rise in the near future because of global population aging. The increasing survival rates related to underlying diseases such as hypertension, coronary heart disease, and heart failure are closely associated with rising prevalence of AF.^[4,5] Globally, AF treatment constitutes a serious medical burden. It is estimated to amount to 1% of the National Health Service budget in the United Kingdom^[6] and $16 to $26 billion of annual US healthcare expenditure.^[7] AF is predicted to affect 16 million people in the US and 16 to 17 million people in Europe by 2050.^[8]

Thermal ablation of myocardial tissue is currently a major clinical treatment for AF. It involves destroying local myocardial tissue to isolate the faulty electrical signals causing the arrhythmia in the local area. Ablation is usually performed by heating local tissue with RF electrical current, creating conduction-blocking lesions that stop AF. The introduction of RF catheter ablation in the field of cardiac electrophysiology can be traced back to 1987. However, the problems associated with RF ablation (RFA) include: (1) high recurrence rate (as tissue cooling by arterial flow may limit transmural and continuous lesion formation),^[9] (2) tissue loss beyond the targeted tissue (non-uniform ablated region),^[10] (3) charring that hinders heat conduction,^[11] (4) “steam pops” that can cause cardiac perforation,^[12] and (5) lengthy ablation procedure.^[13] Additionally, there are small risks of certain complications due to the thermal side effects. In a nationwide US study involving 93,801 RFA procedures from 2000 to 2010, the overall incidence of perioperative complications was 6.29%, and in-hospital mortality was 0.46%.^[14] The most frequent and severe complications of RFA are: (1) stroke (due to thrombosis formation),^[15] (2) pulmonary vein (PV) stenosis,^[16] (3) cardiac tamponade,^[17] (4) coronary artery injury,^[18] (5) phrenic nerve injury,^[19] and (6) atrio-esophageal fistula.^[20]

CB ablation has emerged as a valid alternative in recent years. In one study, CB ablation reduced the incidence of AF recurrence compared to RFA, and this result was consistent across different study designs and AF types.^[21] CB ablation was associated with a significantly higher rate of phrenic nerve palsy, but lower rates of pericardial effusion, cardiac tamponade, and vascular complications than RFA. Therefore, considering its efficacy/safety profile and short procedural time, CB ablation is the preferable procedure for the first AF ablation.^[22]

Recently, experts’ opinions and published data have suggested that combination treatment involving surgical and catheter ablation may provide even better results. One of the most invaluable benefits of surgical ablation is the possibility of concomitant occlusion of the left atrial (LA) appendage.^[23] The availability of cryosurgical techniques has revolutionized the surgical treatment of AF. By using modern cryosurgical devices and adhering to the technical principles described, surgeons can now perform surgical procedures for AF that are quicker, safer, and as effective as the standard Maze-III/IV procedure.^[24]

History of irreversible electroporation (IRE) and its application in cardiology

IRE is a biophysical phenomenon in which the cell membrane permeability to ions and molecules increases significantly in response to an externally applied electric field.^[25] The method was termed electroporation which suggests “creating pores electrically”. When the electric field is of relatively strong intensity, it may induce persistent changes in membrane permeability, leading to an irreversible breakdown of membrane structure and function and ultimately to cell death. This process is called IRE and is an emerging non-thermal modality that is approved by the US Food and Drug Administration (FDA) for clinical tumor ablation. Owing to this non-thermal mechanism, IRE ablation has many advantages compared to conventional RFA. A pilot study of IRE ablation during open-chest surgery in pigs by Lavee et al^[26] in 2007 found that a bipolar 40-mm parallel clamp led to transmural lesions, with a sharp demarcation between the ablated and normal tissues. In Table 1,^[26–37] we detail the significant historical IRE ablation-related events that have made this innovative modality safer and more effective for clinical application, avoiding the multiple severe complications found in RFA. In Figure 1, we summarize these remarkable discoveries from 2007 to 2021.

Table 1 - Selected historical events involving IRE cardiac ablation since 2007.

Study (year)	Experimental model (number)	Electrode configuration and delivery method	IRE protocol parameter	Effects	Follow-up period	Significance
Lavee (2007)[26]	Porcine (n = 5), open chest	Bipolar 40-mm parallel clamp electrodes, Epi_Delv	PN = 8, 16, 32,PD = 100 μs,PRI = 200 ms,PA = 1.5∼2 kV	Transmural lesions (n = 10) with mean 0.9 cm depth, 3.0–3.5 cm length, created in 1–4 s	24 hours	First reported successful atrial transmural IRE ablation in vivo, with clear ablated/unablated boundary.
Hong (2009)[27]	Ovine (n = 4), open chest	Bipolar 53-mm linear clamp electrodes, Epi_Delv	3 or 5 trains of pulses, each train with PN = 10∼40,PD = 100∼400 μs, PRI = 200∼1000 ms, PA = 0.78 kV	100% transmural lesions in SVC and IVC (n = 14), and high degree of transmural in RAA, LAA, and PVs (n = 19)	Not stated	High degree of transmural lesions in different locations in atria. Additionally, esophageal lesions were created showing that IRE ablation is selective (epithelial tissue was kept alive while target muscular tissue was destroyed).
Wittkampf (2011)[28]	Porcine (n = 10), closed chest	Circular 20-mm-diameter with 10-pole 2-mm electrodes, Endo_Delv	200-J DC pulses, PD = a few ms, DC applied between a skin patch and the multipolar circular electrode	Sharp lesions (maximum depth = 3.5 mm) created inside PV ostia with reduced PV potentials	3 weeks	First reported safe and successful low-energy IRE ablation in a sensitive environment (PV ostia). PV angiograms did not show stenosis during the short follow-up.
Wittkampf (2012)[29]	Porcine (n = 10), open chest	Circular 20-mm-diameter with 10-pole 2-mm electrodes, and circular 20-mm-diameter single ring electrode, Epi_Delv	50/100/200-J DC pulse, DC applied between a skin patch and the multipolar circular electrode	Continuous lesions (n = 5) created with median width (5.3 ± 3.0) mm and median depth (5.2 ± 1.2) mm created by 200 J via devices D which contained 2 mm long electrode	3 weeks	Proved lesion depth increased with pulse magnitude in a blood-myocardial tissue environment. Continuous and deep lesions for PVI could be created by a single 200-J pulse of a few ms in duration.
Neven (2014)[30]	Porcine (n = 5), open chest	Linear electrode in suction cup, Epi_Delv	30/100/300-J DC pulses, PD = 6 ms, DC applied between a skin patch and the linear electrode	Lesions with mean depth of (3.2 ± 0.7), (6.3 ± 1.8), and (8.0 ± 1.5) mm, mean width of (10.1 ± 0.8), (15.1 ± 1.5), and (17.1 ± 1.3) mm, transmural rate was 25%, 100%, and 100% for 30, 100, and 300-J epicardial ablation, respectively	3 months	No permanent coronary artery damage. No arterial luminal narrowing at 3 months.
van Driel (2015)[31]	Porcine (n = 20), open and closed chest	Circular 20-mm-diameter with 8-pole 4-mm electrodes, and circular 18-mm-diameter with 8-pole 2-mm electrodes, Endo_Delv	200-J DC pulses, DC applied between a skin patch and the multipolar circular electrode	Transmural lesions after firm contact with the inner superior vena cava wall	3–13 weeks	No phrenic nerve damage.
Xie (2015)[32]	Rabbit (n = 12), isolated heart	Parallel penetrating needle electrodes spaced 2–4-mm, Epi_Delv	PN = 6, 20,PD = 350 ns,PRI = 300 ms, 1 s,PA = 2.3 kV	Transmural lesions verified based on histological and optical mapping data	No	First report of transmural lesions created by nanosecond pulsed electric field.
Neven (2017)[33]	Porcine (n = 8), open chest	Suction cup containing single 35 × 6 mm linear electrode, Epi_Delv	100/200-J DC pulses, PD = 6 ms, DC applied between a skin patch and the linear electrode	IRE ablation was tested purposely on anterior esophageal adventitia	2 months	No esophageal ulceration or fistula.
Reddy (2018)[34]	Human (n = 22), open and closed chest	Bipolar 12-F over-the-wire catheter with 5 splines each containing 4 separate electrodes for Endo_Delv, and bipolar linear closed ring catheter with multiple electrodes for Epi_Delv	Mean 78 J per ablation for Endo_Delv and 1146 J per ablation for Epi_Delv, over a few-seconds train of multiple ms pulses, PA = 0.9∼2.5 kV	15 patients out of 15 are PV isolated with (3.26 ± 0.5) mm lesions from endovascular delivery, and 6 patients out of 7 from epicardial delivery	1 month	First-in-human clinical application of IRE ablation for acute PVI, for both cardiac surgery and catheter-based ablation.
van Es (2019)[35]	Porcine (n = 5), open chest	Suction cup containing single 35 × 6 mm linear electrode, Epi_Delv	200-J pulses, PN = 10, PD = 2 ms, PRI = 400 ms, frequency = 167 kHz	Asymmetric HF(aHF) Lesion depth (2.3 ± 1.0) mm, Symmetrical HF (sHF) lesion depth (1.7 ± 0.8) mm.	3 weeks	aHF IRE ablation avoids skeletal muscle contractions and gas bubble formation.aHF pulses can also create deeper lesions than sHF pulses of the same energy in cardiac tissue.
Reddy (2019)[36]	Human (n = 81), closed chest	Bipolar 12-F over-the-wire catheter with 5 splines each containing 4 separate electrodes, Endo_Delv	Over a few-seconds train of multiple ms pulses, PA = 0.9∼2.0 kV	All PVs were isolated by monophasic (n = 15) or biphasic (n = 66) approaches. With successive IRE refinement, durability improved from 18% to 100%	120 days	In 81 patients, 100% of all PVs were acutely isolated and no additional primary complications such as stroke, phrenic nerve injury, PV stenosis, or esophageal injury occurred over the 120-day median follow-up.
Reddy (2021)[37]	Human (n = 121), closed chest	Bipolar 12-F over-the-wire catheter with 5 splines each containing 4 separate electrodes, Endo_Delv	Over a few-seconds train of multiple ms pulses, PA = 0.9∼2.0 kV	All PVs were isolated by monophasic (n = 15) or biphasic (n = 106) approaches. With successive IRE refinement, PVI success 100%	1 year	1-year Kaplan-Meier estimates for freedom from any atrial arrhythmia for the entire cohort and for the optimized biphasic energy PEF ablation waveform cohort were (78.5% ± 3.8%) and (84.5% ± 5.4%), respectively.

DC: Direct current; Endo_Delv: Endovascular IRE delivery; Epi_Delv: Epicardial IRE delivery; IRE: Irreversible electroporation; IVC: Inferior vena cava; LAA: Left atrial appendage; PA: Pulse amplitude; PD: Pulse duration; PEF: Pulsed electric field; PN: Pulse number; PRI: Pulse repetition interval (or pulse cycle length); PV: Pulmonary vein; PVI: Pulmonary vein isolation; RAA: Right atrial appendage; SVC: Superior vena vein.

Figure 1: Timeline summarizing the remarkable historical events involving IRE ablation for treating atrial fibrillation. IRE modalities, comprising direct current (DC), alternative current (AC), and pulsed electric field (PEF) ablation, are highlighted in yellow. IRE: Irreversible electroporation; PV: Pulmonary vein; PVI: Pulmonary vein isolation.

IRE modalities

Depending on the protocol, IRE can be achieved with various modalities: direct current (DC), alternative current (AC), pulsed electric field (PEF), or any combination of these.^[38] DC catheter ablation was routinely applied between 1980 and 1990 for the ablation of the bundle of His and ventricular tachycardia.^[28] Despite the high success rate, DC ablation was quickly replaced by the more elegant RFA technique that appeared around 1990.^[39] The main disadvantages of DC monophasic pulses ablation are that they can lead to potentially painful nerve and muscle capture, arcing, metallic release in solution, and bubble stream formation, also due to electrolysis without arcing, and require complete anesthesia during ablation.^[40] In 2011, Wittkampf et al^[28] successfully used low-energy DC ablation to create myocardial lesions, avoiding the harmful side effects of standard DC ablation such as sparking and explosions. The DC shocks (pulse duration of a few milliseconds) were applied between a catheter. The electrode was positioned on the target myocardium with an external ground pad.

There are not as many reports of AC ablation as there are of DC or PEF. A pilot study of high-frequency AC ablation by van Es et al^[35] in 2019 found that asymmetric high-frequency ablation in pig hearts successfully created myocardial lesions without skeletal muscle contractions or bubble formation. PEF ablation is an adaptation of DC ablation, in that it is a more controlled low-energy form of ablation delivered over a few seconds across multiple electrodes without the use of an external ground pad.^[36] PEF consists of multiple short DC pulses with a pulse duration of microseconds or nanoseconds. The local electric field created in a bipolar fashion by PEF leads to the reorganization of the lipid structure of cardiomyocyte membranes, resulting in nanoscale pore formation. These pores increase cell membrane permeability and thus cell leakage, finally leading to cell death. This phenomenon subsequently results in either immediate necrosis or delayed apoptotic cell death.^[41–44] Catheter-based pulmonary vein isolation (PVI) with pentaspline PEF ablation catheter creates chronic PV antral isolation areas as encompassing as thermal energy ablation.^[45]

IRE protocol parameters

IRE protocols include many parameters that can be modified. These parameters include biphasic or monophasic pulses, voltage amplitude (hundreds to thousands of volts), pulse duration (nanoseconds to milliseconds), pulse repetition interval, interphase interval, number of pulses in the train (single pulse to hundreds), number of trains, and bipolar or unipolar electrode delivery. The characteristics of each pulse that can vary include pulse rise time, pulse fall time, overshoot and ringing, and undershoot.^[46] Each of these parameters, combined with others, can significantly influence the efficiency of IRE ablation and result in more/less tissue destruction.

Tissue specificity

Thermal energies related to high-power, short-duration RFA, RF balloon devices, ultra-low cryoablation, laser, ultrasound, microwave, and X-ray ablate tissues indiscriminately.^[47] By contrast, IRE ablation (either DC or PEF), because of its non-thermal ablation nature, can ablate tissues with unique tissue selectivity. This is because the myocardium has a lower threshold (400 V/cm) than the surrounding tissues, therefore IRE will first induce cardiomyocyte necrosis.^[34] If IRE protocols are precisely controlled, this can limit collateral damage to periatrial non-target tissues such as the esophagus and phrenic nerves, coronary arteries, and PVs.^{[29–31,33,34,36]} This underlying mechanism of IRE ablation offers a degree of tissue-specific safety: tissues with higher thresholds in an ablative zone can be spared, avoiding clinical complications inherent to RFA (refer to Table 2, which lists all aspects for comparison).

Table 2 - Comparison of RFA and IRE ablation.

Items	RFA	IRE ablation
Clinical complications
PV stenosis	Yes	No∗
Coronary artery damage	Yes	No∗
Phrenic nerve injury	Yes	No∗
Atrio-esophageal fistula	Yes	No∗
Stroke (due to thrombosis)	Yes	No
Problems associated with thermal effect
Lengthy procedure	Yes	No
High recurrence rate	Yes	No∗
Tissue loss beyond target tissue	Yes	No
Charring	Yes	No
“Steam pop”	Yes	No
Tissue-catheter contact force sensing required	Yes	No
Possible complications
Bubbles	No	Depends on external factors
Arcing	No	Depends on external factors
Muscle contractions	No	Depends on external factors
Other factors
Apoptosis	No	Yes
Cardiac fiber orientation dependence	No	Depends on external factors
Areas of reversibility	Yes	Depends on external factors
Safety and efficacy can be achieved concurrently	Yes	No

^∗Based on short-term data (long-term data are not available). In summary, for RFA, there is a zero-sum relationship between safety and efficacy (greater efficacy is achieved with more ablation but at the expense of safety, and vice versa). However, for IRE ablation, the safety and efficacy relationship is a win-win relationship (they can be achieved at the same time). IRE: Irreversible electroporation; PV: Pulmonary vein; RFA: Radiofrequency ablation.

No evidence of PV stenosis

Stenosis of the PV is a serious and potentially life-threatening complication of RFA caused by hyperthermia that induces contraction of the vein wall. The non-thermal mechanism of IRE ablation leads to a much lower propensity for collateral damage compared to procedures using thermal energy sources.^[48] In an animal study, the PV diameter was maintained at 3 weeks after IRE DC ablation, while RFA decreased the PV diameter at 3 weeks compared to that in the control group.^[28] It was also confirmed in patients that there was no PV stenosis at a 1-month follow-up visit after IRE ablation.^[34]

No evidence of coronary artery damage

Thermal ablation induces heat damage to all tissue near the ablation site, including nearby coronary arteries, which can pose a serious risk. The heat may not only coagulate blood inside the vessel, causing thrombus, but also cause intimal hyperplasia and shrinkage of the collagen fibers in the arterial wall that could lead to vessel stenosis and subsequent infarction of the related region. In contrast, IRE ablates the myocardium but selectively spares the arteries themselves entirely. This selective ablation feature can be used in oncological IRE treatment as solid tumors are ablated, while nearby vital structures such as blood vessels and nerves are relatively spared, given that they are relatively resistant to IRE PEF ablation.^[9] Neven et al^[49] proved that 3 months after epicardial IRE, luminal coronary artery diameter remained unaffected. IRE can create deep lesions and is a safe modality for catheter ablation on or near coronary arteries.

No evidence of phrenic nerve damage

The phrenic nerves are located very close to the superior vena cava, right PVs, and LA appendage. Phrenic nerve palsy is another complication of PVI by using thermal ablation.^[19] In 2015, van Driel et al^[31] found no evidence of right phrenic nerve damage after intentionally targeted ablation using circular 200-J DC applications with good tissue contact inside the superior vena cava wall. Occasionally, transient nerve palsy was induced, but normal pacing was monitored from the superior vena cava 30 minutes and 3 to 13 weeks after ablation.

No evidence of atrio-esophageal fistula

Atrio-esophageal fistula is another rare but devastating complication of thermal LA ablation. It occurs due to the anatomical location of the LA near the esophagus. Excess heat conduction from the LA to the esophagus can damage esophageal tissue, leading to fistula.^[20] IRE ablation involves tissue-specific thresholds, thereby facilitating preferential ablation of only the myocardium, with relative sparing of collateral tissues such as the esophagus. In 2017, Neven et al^[33] reported that the architecture of the esophageal tissue remained intact after 2 months after irreversible electroporation. In the novel porcine model of esophageal injury, biphasic PEF ablation induced no chronic histopathologic esophageal changes, while RFA demonstrated a spectrum of esophageal lesions including fistula and deep esophageal ulcers and abscesses.^[50]

No evidence of stroke (due to thrombosis)

Thermal ablation carried out in very close proximity to a coronary artery can cause thrombosis and occlusion, as the heat may coagulate blood inside the vessel. Owing to the lack of an obvious temperature increase associated with IRE ablation, the risk of heat-induced thrombus formation is low.

Procedure time

It is time-consuming to use thermal ablation to achieve measurable effects and appropriate temperature gradients. By contrast, IRE is almost instantaneous, occurring in a few seconds [Table 1]. Thus, using IRE ablation dramatically reduces the cumulative time. In 2018, Reddy et al^[34] reported that the IRE procedure took (67.0 ± 10.5) minutes (57 PVs in 15 patients). However, in a 2016 study by Boersma et al^[51] involving PVI, RFA took (96 ± 36) minutes (61 patients) and using a standard focal irrigated catheter took (166 ± 46) minutes (59 patients).

Recurrence rate

AF recurrence after RFA was common in patients with persistent AF in studies by Cappato et al,^[52] Callans et al,^[53] and Verma et al.^[54] This recurrence of PV-LA conduction is associated with the presence of gaps in incomplete lesions, created by RF energy, between the PV and LA. Durable thermal ablation efficacy for AF remains a major challenge in the long term. Depending on the ablation strategy, success rates range from 60% to 80% for paroxysmal AF, and from 50% to 60% for persistent AF.^[55] Among the ablation strategies, PVI constitutes the cornerstone of the thermal catheter ablation strategy, by creating a circumferential ring of non-conductive lesions around the individual PV from the LA.^[56] The reasons for AF recurrence (ie, recurrence of PV-LA electrical conduction) in the long term after ablation can be classified into 4 possible groups: (1) recovery of PV-LA conduction that was previously ablated, (2) presence of triggers ostial to the PVI site, (3) development of new triggers in non-ablated PVs, or (4) development of new triggers outside the PVs.^[57]

Generally, short-term PVI can be achieved even by incomplete lesion formation that causes temporary electric uncoupling but not irreversible cell death (reversible tissue injury) at ≤3.5 mm from the lesion boundary.^[58] Ablation involving a gap of ≤1.4 mm can still cause a local conduction block, and the gap can be up to 4 mm in abnormal tissue.^[59] These lesions with gaps “mask” the need for permanent complete lesion formation for permanent PVI. Formation of lesion gaps are due to insufficient heat, which is carried away by convective cooling involving nearby arterial blood flow. Over 1 to 4 weeks, the normal conduction of live tissue at the boundary of the transmural lesion recovers, and PV-LA conduction gradually recovers over several weeks or months as the tissue injury heals.^[58] Therefore, permanent PVI should be considered as the main goal of AF treatment to avoid AF recurrence. However, the verification of the AF recurrence rate after IRE ablation is still in the early stage, owing to its short history of use. Reddy et al^[36] are the pioneers of IRE PEF ablation for treating AF: the rate of durable PVI at 3 months improved from 18% for the initial monophasic waveform to 100% (with all PVs isolated) for the biphasic waveform with successive “waveform refinements”. The details of the successive waveform refinements were not reported, but it appears that bipolar basket electrode configuration without a skin patch is an effective delivery approach for PEF ablation.

In 3 multicenter studies (IMPULSE, PEFCAT, and PEFCAT II), patients with paroxysmal AF underwent PVI using a basket or flower PEF ablation catheter. Invasive remapping was performed at 2 to 3 months, and reconnected PVs were reisolated with PEF ablation (with the optimized biphasic energy PEF ablation waveform) or RFA. Arrhythmia recurrence was assessed over a 1-year follow-up period, and PVI was achieved in 100% of PVs with PEF ablation, with durable PVI in 96.0% of PVs treated with PEF ablation. The 1-year Kaplan-Meier estimate for freedom from any atrial arrhythmia for the optimized biphasic energy PEF ablation waveform cohort was 84.5% ± 5.4%.^[37]

Despite this, the complications, which are currently thought to be rare, may actually occur at higher rates than previously thought, as was the case when other new technologies were previously evaluated (eg, cryo, laser, and focused ultrasound).^[60–62] Therefore, it is inappropriate to state that each of the risks is definitively lower for IRE ablation than RFA. The true risks are unknown, although the limited preliminary data are promising.^[63]

Tissue loss beyond the target tissue

Based on the time-dependent conductive thermal mechanism of RFA, the target zone temperature must involve the following to effectively create transmural lesions: (1) the appropriate peak temperature must be reached at the proximal tissue-electrode interface and (2) the peak temperature must spread from the proximal to distal tissues to destroy the tissue transmurally from the endocardium to the epicardium. In general, the cross-sectional area of lesions is elliptical, that is, it is larger proximally and smaller distally. To ensure sufficient distal damage for a transmural lesion, more proximal tissue has to be sacrificed. Another extra loss of tissue is necessitated because of convective cooling by the arterial blood flow that will carry some heat away, limiting distal lesion formation. To create adequate distal lesions, extra proximal tissue must be sacrificed. In contrast, IRE ablation is not based on a time-dependent thermal mechanism. By selecting the optimized IRE protocol parameters and electrode configuration, a well-controlled ablation region can be expected, with a clear ablated-unablated boundary.^[64]

Charring and “steam pop”

During RFA, charring hinders heat conduction, thus affecting the efficacy of lesion formation. Additionally, if tissue temperatures exceed the catheter tip temperature and go beyond 100°C, steam explosions can occur, which are known as “steam pops”.^[65] They can cause superficial craters or deep tissue tears, which can lead to cardiac perforation and tamponade.^[59] Fortunately, due to the nature of PEF energy, the non-thermal IRE PEF ablation mechanism eliminates high-temperature phenomena that occur during RFA such as catheter tip charring and “steam pops”.^[66,67]

Tissue-catheter contact dependence

For RFA, achieving good tissue-catheter contact is critical to produce adequate lesion depth and width. Insufficient contact between the tissue and electrode may lead to shallow and discontinuous lesions that may not be acutely effective.^[38] Insufficient contact force (between the electrode tip and the myocardium wall) can result in inadequate lesion formation and higher rates of PV re-connection, but excessive contact force can result in complications such as perforation.^[68] Therefore, continuous monitoring of the contact force would be expected to maximize the ablation efficacy of RFA and improve its safety. In contrast, IRE relies on tissue proximity and is not solely contact dependent, because it depends on the lethal electric field strength. Although the relationship between the electric field and the applied external voltage is complicated in the heterogeneous tissue, the local electric field still equals local voltage divided by distance at the specific point. If the local voltage increases and the distance is constant, the local electric field increases. The creation of an electric field, rather than contact-dependent lesions, is an attractive feature of cardiac IRE ablation, which does not require good contact between the tissue and catheter tip, eliminating the need for contact force sensing and the dependence on tissue temperatures that cause cell death.

Limitations associated with IRE ablation

Owing to the short history of development of IRE ablation as an AF treatment, the clinical limitations have not yet been fully explored. However, the following is a list of a few of the most concerning limitations.

Involuntary muscle contractions

Conventional IRE DC ablation, which was routinely applied between 1980 and 1990 for the ablation of the bundle of His and ventricular tachycardia, can lead to significant involuntary skeletal muscle contractions and pain, necessitating the use of complete anesthesia.^[28] The DC is delivered in a unipolar fashion, involving an active electrode coupled with a reference dispersal electrode patch on the skin. By contrast, bipolar energy delivered between local electrodes may confine the electric field, resulting in effective local tissue ablation without significant muscle contractions. Depending on the delivered energy, complete anesthesia may or may not be required. In 2019, Reddy et al^[36] reported that, in 15 out of 81 patients undergoing monophasic PEF ablation under general anesthesia, skeletal muscle contractions were not observed. In a group of 106 patients undergoing biphasic procedures under conscious sedation, only mild degrees of muscle activation that were well tolerated were observed, and there was occasional transient intra-procedural cough.^[37]

Bubble formation and arcing

Standard DC causes electrolysis on the metal electrode surfaces in the blood pool, thus producing micro bubbles containing the gaseous products of electrolysis. Arcing is another phenomenon that occurs when a critical electric current density between the electrode and surrounding blood is reached; the resulting flash or arc leads to a transient temperature rise up to a thousand within the bubble.^[69] In 2011, Wittkampf et al^[28] explored low-energy DC with a modified electrode configuration and reported the creation of myocardial lesions without arcing and only small gas bubbles that adhered to the metal electrode surface and disappeared within a few seconds. Furthermore, using a low-energy PEF protocol, in 2019, Reddy et al^[36] reported no arcing or significant visible bubble formation.

Cardiac fiber orientation dependence

The local electric field distribution (electric current density divided by conductivity tensor) determines the effective ablated cardiac tissue region. Due to the anisotropic orientation of cardiac fibers, a lethal electric field cannot create consistent lesions, according to a numerical model reported by Xie and Zemlin^[70] in 2016. In addition, the existence of fibrosis and blood vessels can make the myocardium conductivity even more heterogeneous. Moreover, the electrical conductivity of infarcted myocardial tissue is higher than that of healthy myocardial tissue.^[71] Therefore, IRE protocols must be tailored according to the target cardiac tissue,^[72] to make the ablation lesion more consistent. Unfortunately, due to the complicated geometry of the heart, there are very few studies to investigate how varying conductivity and protocol modifications affect heterogeneous ablation lesions.

Areas of reversibility

RFA-induced non-conductive lesions (including complete and incomplete) originate from the combination of irreversible and reversible atrial thermal injury. An incomplete lesion involving reversible damage can cause temporary electric uncoupling but not complete cell death at ≤3.5 mm from the lesion boundary (dependent on tissue temperature and tissue geometry), which increases the risk of long-term AF recurrence.^[58] Owing to the short history of use of IRE ablation, there is a lack of evidence on whether it can cause irreversible injury, and it is not clear how far away from the lesion boundary temporary non-conductivity can occur.

Conclusion and future perspectives

IRE ablation has significant potential to become an important ablation modality in both invasive and interventional cardiology. In the field of AF treatment, the unique ability of IRE to induce a tissue-selective and well-controlled ablation zone with no thermal damage offers important advantages, as its non-thermal nature reduces the risk of complications, such as PV stenosis, coronary artery damage, phrenic nerve injury, and atrio-esophageal fistula. The absence of a large thermal effect also decreases the ablation procedure duration, lowers the recurrence rate, avoids charring, and may allow direct treatment of vessels, where the muscular cell layer is destroyed but the structural integrity of the vessel is preserved, as has been demonstrated in previous studies.^[64,73] A high level of safety has been demonstrated due to IRE's tissue selectivity, which spares nerves, coronary arteries, esophageal muscle, phrenic nerves, and PVs, while destroying the myocardium. However, the mechanism underlying the lower IRE electric field strength threshold of cardiac tissue compared to surrounding tissues is not completely clear. It may be related to cell size, cardiac fiber orientation, membrane characteristics, and sensitivity to non-specific cation entry.^[46] With axons connected, only parts of nearby collateral membranes are exposed to the IRE electric field, but these parts may be too small for the effects to be lethal.^[38] It has been confirmed in animal and human studies that IRE ablation does not have many of the short-term complications of RFA.^[74,75] However, long-term follow-up (using computed tomography/magnetic resonance imaging) to assess long-term complications such as PV stenosis, coronary artery damage, phrenic nerve injury, and atrio-esophageal fistula is required before IRE ablation technology is applied clinically.^[37]

In terms of efficacy, efforts should be made to develop IRE ablation strategies that achieve more durable lesions. Ablation strategies to ensure permanent PVI (to prevent recurrent PV-LA conduction) may have the greatest impact on reducing AF recurrence. Although Reddy et al^[36] reported that IRE PEF ablation with waveform refinement improved the PVI durability at 3 months from 18% to 100%, provocative testing with adenosine or isoproterenol was not performed, and longer-term research needs to be conducted to ensure that the number of patients with permanent PVI is increased, which will result in a lower AF recurrence rate. Until recently, IRE ablation catheters have been used for PV ablation, but new IRE ablation catheters can be used for non-PV ablation, including posterior wall isolation and lines of conduction block; moreover, a study found that Purkinje tissue can be ablated using IRE without any evidence of underlying myocardial damage.^[76]

Although pre-clinical studies indicate excellent efficacy and safety profiles for IRE ablation, it has several disadvantages, such as the need for general anesthesia and neuromuscular paralysis to avoid skeletal muscle contractions; furthermore, arcing and bubble formation need to be avoided. Thus, an optimized IRE ablation protocol that avoids muscle contraction, pain, arcing, and bubble formation is highly desired. Additionally, the features of cardiac fiber orientation dependence and areas of reversibility need to be investigated, to fully understand the whole picture of how to exploit this promising technique that may replace conventional RFA.

IRE PEF ablation is a new technique for treating AF. Animal and clinical studies have confirmed its feasibility and safety.^[64,73] The limited preliminary data are promising. The technique has strong tissue specificity and a short ablation time, which can effectively reduce perioperative complications and shorten the operation time. The limited data currently available indicate that the injury caused by IRE PEF ablation is tissue specific, and the risks of complications such as PV stenosis, nerve injury, and atrial esophageal fistula are low. However, at present, clinical research on this technique is rare, the sample sizes are small, the follow-up periods are short, and it is necessary to carry out surgery under general anesthesia, which currently limits its extensive clinical use. There is still a lack of robust data on the efficacy of IRE ablation for AF compared to conventional RFA/CB ablation, which needs to be further investigated using randomized controlled trials.

Funding

This work was supported by National S&T Major Project (2018ZX10301201), Chinese Natural Science Foundation Grants (82027803), Science and Technology Development Special Fund of Shanghai Health and Family Planning Commission (ZK2019B25), Plateau Discipline Construction Fund of Pudong New Area Health and Family Planning Commission (PWYgy2018-03), Chinese Academy of Engineering (2019-ZD-6-01), and Zhejiang University Education Foundation (2020XGZX063).

Conflicts of Interest

None.