In developed countries, the number of atrial fibrillation (AF) catheter ablation procedures increases every year. Continuous scientific and technological innovation in the area of AF ablation has broadened the spectrum of therapeutic options for patients with AF, and it is likely to be a contributing factor in making AF ablation a practice no longer restricted to a few operators with greater technical skills in high- volume electrophysiological centres. Nevertheless, AF ablation remains a technically challenging procedure for which new technologies allow new approaches, while new approaches require new technological improvements. The main purpose of developing technologies is to improve AF ablation efficacy by means of better imaging definition, more effective ablation catheters, remote navigation systems and automated algorithms aimed at characterising atrial substrate. The other major concern of new technologies pertains to increasing attention being paid to the safety profile.
Newer imaging technologies have been developed to improve our knowledge of true cardiac anatomy, thus orientating the operator in navigating the ablation catheter through the complex anatomy of the left atrium (LA) and the pulmonary veins (PVs) and locating the lesion sets. Several imaging modalities are available, each with its own set of advantages and disadvantages, but the best imaging rendering arises from the combination of two or more strategies.
Pulmonary Vein Angiography
Visualisation of PVs was originally obtained by retrograde venous angiography performed with a long sheath positioned sequentially into the four PVs. This approach is quite safe and economical and provides information about the anatomy of the veins and ostium, but it does not allow a definite detection of additional veins or other structures or supply realtime guidance for catheter navigation. An important indication for this technique is the assessment of PV occlusion during balloon-based ablation.
In order to obtain better anatomical definition, the use of contrast- enhanced rotational angiography has recently been introduced. Contrast medium is injected into the pulmonary artery, right atrium (RA) or LA; the fluoroscopy C-arm is then rapidly rotated around the patient and images are acquired throughout the rotation to generate 3D volumetric anatomical rendering of the LA-PVs, which can be superimposed onto the fluoroscopic projections or integrated into an electroanatomical mapping (EAM) system.1–3 A new software version allows for the inner surface to be seen (endoscopic view).2 The rotational angiography may help overcome some of the limitations of a pre-acquired computed tomography (CT) or magnetic resonance imaging (MRI) scan. Since the images are obtained at the time of theprocedure, they are more likely to represent the true anatomy than a remotely acquired image because of factors such as the patient’s intravascular fluid imbalance, heart rhythm and heart rate. This technique also reduces the financial and administrative burden of scheduling adjunctive and expensive imaging studies. In the case of patient movement, the geometry needs to be fully re-acquired; this can be promptly managed, but the consistent iodinated contrast agent load and radiation dose typically preclude the use of rotational angiography more than twice in a study. Another important limitation is the incomplete opacification of the LA appendage (LAA), which limits the visualisation of this critical structure. Future directions could include methods to update geometry or to perform nearly realtime 3D imaging.
3D Transoesophageal Echocardiography
Recently, realtime 3D transoesophageal echocardiography (TEE) has become available for clinical practice, offering clear and detailed rendering of the cardiac anatomy. The 3D TEE probe (Matrix 3DTEE, Philips, Inc., Andover, MA) allows for both 2D and 3D realtime imaging of the LA-PVs. One strong point of 3D TEE is the excellent visualisation of the interatrial septum and LAA.4 A small study implied that 3D TEE guidance might provide safer trans-septal puncture in patients with unusual anatomy,5 as it offers the benefit of recognising some shapes of the atrial septum that are not well characterised by 2D TEE. Despite these promising features, post-acquisition image processing is necessary and time-consuming. Moreover, general anesthaesia along with endotracheal intubation would be mandatory.
Electric phased-array intracardiac echography (ICE) transducer catheters, which are capable of providing 2D, Doppler and colour-flow images,6–9 have been extensively used in order to guide trans-septal catheterisation, help define the anatomy of the LA-PVs and visualise critical structures, provide online awareness of catheter position and lesion deployment, monitor PV ostial narrowing during and after ablation and for early detection and prevention of complications such as steam pops, thrombosis and oesophageal damage. One significant recent advance is the integration of realtime ICE and EAM9 through the incorporation of a magnetic location sensor into the ICE catheter (Soundstar, Biosense-Webster, Inc., Diamond Bar, CA). By tracing endocardial surface contours on the imported 2D ultrasound images, 3D anatomical reconstructions of the LA-PVs can be created. Ultrasound-derived anatomical shells can be co-registered with pre-operative CT/MRI or rotational angiography imaging; they can also be integrated with electroanatomical data (CartoSound, Biosense Webster Inc.). Compared with CT and MRI, ICE offers the advantage of showing the realtime detailed anatomy of the cardiac chambers that can be updated multiple times during the procedure. Furthermore, creating a 3D reconstruction without entering the LA may reduce procedural time, enhance the safety of catheter ablation procedures and eliminate geometrical distortion resulting from distension of the tissue. However, this technology bears an important financial burden because of the employment of expensive non-reusable ICE catheters.
Computed Tomography or Magnetic Resonance Imaging
CT scan or MRI image integration can be used to create 3D anatomical reconstructions of the LA, which are then imported into EAM systems (NavX Fusion, St Jude Medical, St Paul, MN,10 and CartoMerge, Biosense Webster, Inc., Diamond Bar, CA11,12) to facilitate the intraprocedural visualisation of the LA-PVs. Image integration maintains the benefits derived from the EAM system (visualisation of the mapping catheter, reduction of fluoroscopy exposure, acquisition of ablation points), and it also overcomes the concern that the LA anatomy reconstructed by the 3D mapping system might not match the patient’s real anatomy; in fact, EAMs alone tend to distort the geometry because of catheter- related distension of the cardiac tissue and non-uniform contact, and they depend greatly on the skill of the operator and cartographer to produce CT-type geometries. CT scan and MRI can be used to determine atrial size, number of PVs or the presence of PV stenosis. One of the major benefits of CT and MRI images is that real patient geometry with good details of veins, appendages and other vital structures is obtained in a short amount of time. A recent randomised, prospective trial addressed the controversial topic of the impact of image integration on the procedure/fluoroscopy time and the outcome of the ablation procedure; in contrast with previous non-randomised studies, the authors reported that image integration using Carto-Merge does not significantly improve the clinical outcome of AF ablation, but shortens the X-ray exposure.13 The main limitation of CT and MRI images is that they are not acquired at the time of the procedure, hence raising concerns in terms of changes in chamber size and shape or oesophagus position during the time interval between the image acquisition and the ablation.
In the future, pre-acquired image integration may be replaced by realtime MRI. Intra-procedural cardiac MRI is an attractive option to improve the safety and efficacy of AF catheter ablation for a number of reasons. First, it is radiation-free. Second, it offers the possibility of myocardial tissue and ablation lesion imaging and visualisation of the surrounding structures. In addition, the ability to obtain images in arbitrary orientations increases the potential for high-quality visualisation of catheters, anatomy and electrode–tissue contact. Finally, the position errors introduced by registering catheter position to pre-acquired images can be largely avoided because both realtime and 3D MRI images are acquired in the same co-ordinate system, and 3D images can be re-acquired during the procedure if needed. The ability to use realtime MRI to carry out electrophysiological procedures has been demonstrated mostly in animal models.14 However, catheter guidance may be limited by electrode heating, current induction, image distortion and electromagnetic signal interference; these hurdles could be overcome by using non- ferromagnetic catheters and modifying MRI acquisition settings. The advances in cardiac MRI have revealed the characterisation of fibrotic areas in persistent AF left atria. Recent studies have demonstrated the value of using MRI to characterise myocardial scars before and after the ablation procedure.15,16 Further research is needed before the true clinical value of MRI-guided AF ablation can be evaluated.
Wide areas of technological innovation are aimed at improving the acute and chronic outcome of AF ablation by obviating the need for a high degree of manual dexterity in performing such an extensive lesion set, which is still the prerogative of greatly experienced operators, and by raising the durability of ablation lesions over time. The first concern is addressed by the design of new ablation catheter configurations, for example balloon or multielectrode catheters, to ease the process of PV engagement and encircling, and the introduction of remote navigation systems, which offer the chance to overcome the strict dependence on operator’s skill. Lesion stability can be ameliorated by the utilisation of alternative sources of energy or with the assistance of contact-force- sensing technology. Given the increasing number of patients with structural heart disease and the expanding indications for non- paroxysmal AF ablation, the necessity for straightforward tools to target critical regions other than the PVs, such as high-frequency and fractionated electrogram areas, is another burgeoning field of interest.
Balloon and multielectrode ablation catheters are used in an attempt to perform PV isolation in a shorter time and with a minimum number of lesions, to standardise its completeness and steadiness and to partly obviate the need for manual dexterity. Large multicentre clinical trials will eventually be necessary to determine the safety of these catheters and their ability to overcome the challenges of inter- and intra-patient variability in PV shape/size, as well as to achieve acceptable outcomes. An important limitation of balloon-based ablation is that it only offers PV isolation and does not address other atrial areas that appear to be important for the perpetuation of non- paroxysmal AF; to target these locations, a switch to conventional catheters in addition to the balloon would be required, thus increasing procedural costs and the risk of complications. Other possible restrictions are the inability to isolate all anatomical variations and the more distal level of isolation, which could lead to an increased rate of phrenic nerve (PN) injury.
The greatest clinical experience with balloon ablation technologies has been that utilising the cryothermal energy (Cryocath Technologies, Inc., Kirkland, Quebec, Canada). The catheter is available in 23 and 28mm-diameter balloon sizes. Once the balloon is inflated at the ostium of the targeted PV and cooled, its surface can isolate the vein within a single lesion. The system has been approved for clinical use in Europe and is currently the subject of two randomised, multicentre trials comparing its safety and efficacy with that of antiarrhythmic drugs and radiofrequency ablation (RFA), respectively. Currently, the efficacy of cryoballoon ablation appears comparable to that of standard RF,17–19 but further investigation is claimed to assess its safety profile. The risk of PN injury seems to be one limitation to this approach; however, PN pacing during ablation of the right superior PV can considerably reduce this risk. When the palsy occurred, it predominantly regressed during follow-up. Minor oesophageal damage has been observed with balloon cryoablation, although no atrio-oesophageal (AE) fistulae have occurred to date.
Visually Guided Laser Balloon Ablation
Clinical trials with a visually guided laser balloon catheter (CardioFocus, Inc., Marlborough, MA) are being conducted in Europe.20 This compliant balloon catheter is equipped with an endoscope that provides direct visualisation of the LA-PV junction after inflation. An optical fibre within the balloon can then deliver an arc of ablative laser energy to the visualised tissue, which appears blanched white if it is in good contact with the balloon surface or red if blood is in between the two surfaces. The two major aspects to the visual guidance potential of this system are the balloon positioning relative to the LA-PV anatomy and aiming the laser arc to target the appropriate locations. Early clinical experience with this technology suggests its feasibility and the ability to achieve PV isolation rates comparable to those obtained with RF energy in patients with highly variable PV shape and size.
‘Hot’ Balloon Ablation
This elastic and compliant balloon ablation catheter (Toray Industries, Inc., Houston, TX) can achieve better thermal contact with irregular LA-PV anatomies for the placement of continuous circumferential antral ablation lesions at the targeted PV. The RF generator delivers a high-frequency current (1.8–13.56MHz) to heat saline within the balloon to a stable temperature ranging between 60 and 75°C. Due to large heat capacity of the blood, only the tissue in direct contact with the balloon is selectively ablated.21 Although the experimental experience with this balloon appears promising, there is currently minimal clinical experience with this device.
Balloon-based High-intensity Focused Ultrasound
The high-intensity focused ultrasound (HiFU) balloon catheter (ProRhythm, Inc., Ronkonkoma, NY) was designed to deliver a forwardly directable circumferential ring of ablation energy and showed the ability to safely deliver energy through blood and electrically isolate veins within a few seconds of ablation. After encouraging initial results, its effectiveness has been offset by the severe complications reported after HiFU treatment (permanent PN palsy, significant PV stenosis, fatal AE fistula). This led to a halt of its clinical use and further development.22
Multielectrode Ablation Catheters
A novel technology that delivers duty-cycled unipolar and bipolar RF energy at 8–10W through anatomically specific multielectrode catheters (Ablation Frontiers, Inc., Carlsbad, CA), as well as a mapper and ablator MESH catheter (Bard Electrophysiology, Lowell, MA), have been developed. These catheters are available in different shapes, according to the targeted area: circular and mesh array shaped catheters are used for PV isolation, while other configurations, such as cross- and Y-shapes, are aimed at linear ablation of the atrial substrate.23–25 The electrodes can ablate individually, selectively or altogether. This technology bears the potential to obviate the need for a double trans-septal puncture and to avoid switching catheters through the sheath, as both high-density mapping and RF ablation can be performed by the same catheter. The ability of some of these catheters to modulate the bipolar/unipolar energy ratio, although achieving transmural lesions, can help avoid collateral damage by reducing the amount of unipolar energy while approaching critical structures such as the oesophagus. A few studies showed good success rates and a favourable safety profile in the ablation of both paroxysmal and long-standing persistent AF in small populations. The unavailability of external irrigation for multielectrode catheters may represent a limitation; however, the possibility of using alternating bipolar/unipolar RF energy with a limited power output minimises the risk of thromboembolic complications and other collateral damage.
The contact force applied to the tissue during RF ablation has been identified as one of the major determinants of lesion volume and depth and procedural safety. Excess force more frequently results in cardiac complications, such as steam pops, coagulum or char formation at the electrode tip, which may give rise to perforation or stroke. Force sensing is a recently developed technology that allows the determination of the realtime contact force at the catheter tip during ablation. Early pre-clinical studies and preliminary results from a clinical trial have shown that force-sensing technology improved catheter contact, ablation lesion size and overall success and complication rates.26,27
Manual catheter manipulation faces important limitations, such as reproducibility, dependence on operator’s skill, catheter stability and operator fatigue, which together can affect the procedural outcome. Radiation exposure of the operator is also an important concern. Remote catheter navigation has emerged in recent years to address these concerns.28–30 Currently available technologies employ either the manipulation of an external magnetic field to guide a floppy catheter (Niobe, Stereotaxis, Inc., St Louis, MO) or the robotic manoeuvring of steerable sheaths to guide a conventional catheter (Sensei Robotic Catheter System, Hansen Medical, Mountain View, CA). Both of these devices have been integrated with EAM systems to store catheter location information for semi-automated re-navigation to regions of interest. These systems translate the operator’s manipulation of a handle into precise movements of the catheter, thus allowing barely accessible regions of the heart to be reached, to create detailed EAMs and precise ablation lesions. Some studies28–30 demonstrated the safety and feasibility of the system for both mapping and ablation, along with a significant reduction of fluoroscopy time. However, randomised clinical trials are still required to determine the comparative efficacy of remote technologies versus manual ablation for the treatment of AF.
Targeting Complex Fractionated Atrial Electrograms
Initially, long linear lesions in both atria were introduced to interrupt potential macro-re-entries and to improve the success rate of catheter ablation of nonparoxysmal AF. However, incomplete line deployment led to an increase in the number of iatrogenic gap-related tachycardias that clinically could be even worse for the patient. Several studies have shown that complex fractionated atrial electrograms (CFAEs) represent a critical substrate for AF initiation and perpetuation and may serve as potential ablation target sites. These electrograms, characterised by short cycle length and complex morphology, likely represent the continuous re-entry of fibrillation, overlap of different wavelets entering the same area at different times or non-uniform wavelet propagation (fibrillatory conduction) of remote drivers. In an attempt to overcome the interobserver and intraobserver variability in the identification of CFAEs, signal- processing software algorithms have been made available for EAM systems, which offer a satisfactory diagnostic accuracy in the detection of CFAEs.31,32 A recent randomised multicentre trial found that CFAE ablation guided by automated mapping software may have an additive benefit over PVI alone, but does not suffice as an ablation strategy in and of itself.
Since catheter ablation of AF is an elective procedure for a generally non-fatal condition, procedure-related complications represent a quite traumatic event for both patients and physicians. The major complication rate reported in the updated worldwide survey by Cappato et al.33 is still 4.9%, even though this is likely to be significantly lower when the procedure is performed by experienced operators. For these reasons, significant effort continues to be expended to the purpose of improving the safety profile.
Pulmonary Vein Stenosis
With the development of 3D mapping systems and the currently extra-ostial ablation strategies, the reported incidence of symptomatic PV stenosis is decreasing.34 Nonetheless, a good registration process with 3D CT/MR imaging is crucial in order to obtain a reliable definition of PV ostia position and avoid inadvertent ablation within the PVs. Cryothermia could also reduce the potential of PV stenosis by minimal endothelial disruption and collagen formation, but this is yet to be proved.
The institution of saline irrigation of the ablation catheter tip has minimised the problem of thromboembolism by preventing excessive electrode temperature elevation and char formation. The duration of the ablation procedure,35 the diameter of the sheath and the number of catheter exchanges required can also contribute to embolic events. In these terms, the need for a 15Fr sheath in combination with an additional mapping catheter remains a handicap for cryoballoon ablation, even though the absence of endothelial disruption may result in a lower thrombogenic burden. Continuation of oral anticoagulation at a therapeutic international normalised ratio at the time of ablation is potentially a better strategy to avoid stroke compared with the use of heparin or enoxaparin.
Cardiac Perforation and Tamponade
Irrigated ablation catheters increase the risk of rapid overheating of the target tissue, culminating in steam pop, cardiac perforation and, eventually, tamponade. Experimental data indicate that the propensity for pops correlates with increased contact force, suggesting a role of recently developed force-sensing technology in reducing these complications. Visually guided trans-septal puncture with newer and 3D imaging techniques could help reduce the occurrence of cardiac tamponade.
The pathogenesis of this rare but potentially life-threatening complication is still incompletely understood.36 The most commonly employed strategy to prevent AE fistula is to limit power output at posterior LA locations in close proximity to the oesophagus. Another approach is the visualisation of the oesophagus during ablation with pre-acquired CT/MR imaging (which has the strong limitation of not providing a realtime guidance), intraprocedural ICE or barium- contrast fluoroscopy, all these options being limited by the absence of any physiological information in terms of the thermal effect on the oesophagus. Other strategies adopted luminal oesophageal devices, such as temperature probes or cooling balloons. The future ability to titrate RF power based on realtime contact force may reduce the risk of oesophageal injury while preserving an effective RF lesion. No study has convincingly proved the efficacy of these strategies in eliminating this complication;37–39 therefore, it is most prudent to adopt multiple simultaneous approaches.
Phrenic Nerve Injury
Balloon-based technology is more prone to the development of PN palsy compared with point-by-point ablation, especially if smaller- diameter balloons are used. The occurrence of PN palsy when using cryoballoon ranges from 2 to 10%. High-output pacing is recommended for identifying the location of the PN before ablation at or near the right superior PV, superior vena cava or proximal LAA roof.17–19
To date, the health hazards associated with ionising radiations exposure in the setting of AF catheter ablation for both the patients and the physicians have not been adequately addressed, but the consistent length of this procedure makes it an interesting topic. The use of EAM systems and robotic navigation proved to significantly reduce fluoroscopy time.40 A future direction towards a further reduction in X-ray exposure could be the introduction of realtime MRI.
The risk of asymptomatic episodes of AF ratifies the importance of maintaining chronic anticoagulation in post-ablation patients.41,42 Despite its proven efficacy, warfarin is often not well tolerated by patients, has a very narrow therapeutic range and carries a high risk of bleeding complications. On the basis of echocardiography and autopsy studies showing that the LAA was the source of thrombi in >90% of patients with non-valvular AF, a percutaneous umbrella-like LAA occlusion device (Watchman, Atritech, Inc., Plymouth, MN, see Figure 2) has been recently marketed. The device is deployed through a trans-septal sheath to exclude the appendage from the circulation. A randomised clinical trial showed that the efficacy of such occlusion devices was not inferior to that of chronic warfarin therapy, at the cost of a clearly higher up-front risk in device implantation, which is predominantly related to LAA trauma and pericardial effusion. An additional European clinical trial is investigating the efficacy of LAA occlusion in patients with a warfarin contraindication. This device may have a significant impact on the quality of life of AF patients who cannot or do not wish to take warfarin.
One of the main hurdles that hamper the comparison of clinical efficacy results among AF ablation studies is the disparity of follow-up strategies. These include the use of symptom-based- only follow-up (which by definition underestimates the incidence of post-ablation asymptomatic episodes of AF), telephone interview, self- palpation of the radial pulse, 24-hour and seven-day Holter electrocardiogram recording, transtelephonic monitoring and implantable loop recorders. A recently published consent statement recommends a minimum of sequential Holter recordings, providing more stringent recommendations for prospective trials (seven-day Holter recording or even implantable loop recorders).43–44 A novel loop recorder (Reveal XT, Medtronic Inc.) is equipped with an algorithm for the discrimination of supraventricular arrhythmias and AF, but to date there are no studies evaluating its efficacy in AF recurrence monitoring after catheter ablation because of factors related to patient acceptance of an implantable device. Cutaneous cardiac rhythm-monitoring patches have recently been introduced into clinical practice. They have the potential to supply a continuous monitoring of the heart rhythm obviating the occurrence of an invasive procedure, but they carry a consistent financial and organisational burden due to the cost of the device itself and the necessity for a centralised system of tracing interpretation by dedicated personnel; furthermore, factors related to interpersonal communication between patients, physicians and professional staff should be taken into account.