Konstantinos Tampakis,Sokratis Pastromas,Alexandros Sykiotis,Stamatina Kampanarou,Georgios Kourgiannidis,Chrysa Pyrpiri,Maria Bousoula,Dimitrios Rozakis,George Andrikopoulos

Abstract Cardiac magnetic resonance (CMR) imaging could enable major advantages when guiding in real-time cardiac electrophysiology procedures offering high-resolution anatomy, arrhythmia substrate, and ablation lesion visualization in the absence of ionizing radiation. Over the last decade, technologies and platforms for performing electrophysiology procedures in a CMR environment have been developed. However, performing procedures outside the conventional fluoroscopic laboratory posed technical, practical and safety concerns. The development of magnetic resonance imaging compatible ablation systems, the recording of high-quality electrograms despite significant electromagnetic interference and reliable methods for catheter visualization and lesion assessment are the main limiting factors. The first human reports, in order to establish a procedural workflow, have rationally focused on the relatively simple typical atrial flutter ablation and have shown that CMR-guided cavotricuspid isthmus ablation represents a valid alternative to conventional ablation. Potential expansion to other more complex arrhythmias, especially ventricular tachycardia and atrial fibrillation, would be of essential impact, taking into consideration the widespread use of substrate-based strategies. Importantly, all limitations need to be solved before application of CMR-guided ablation in a broad clinical setting.

Key Words: Interventional cardiac magnetic resonance;Image-guided ablation;Substrate ablation;Cavotricuspid isthmus;Catheter ablation;Tracking

INTRODUCTION

Cardiovascular magnetic resonance (CMR) has progressively evolved to become an important tool in imaging for cardiac arrhythmias and its implementation is increasingly used[1-3]. By enabling cardiac visualization with augmented temporal and spatial resolution and detailed tissue characterization, CMR imaging identifies both atrial and ventricular arrhythmogenic substrates[1-3]. Accurate scar tissue characterization has been shown to enable prediction of catheter ablation outcome[3], selection of ablation targets for substrate-based procedures[4,5] and identification of gaps in previous ablation lines[5-7]. Moreover, magnetic resonance (MR) imaging may facilitate ablation by providing a detailed anatomical description as pulmonary venous drainage pattern while pre-procedural imaging has also been used for image integration[8,9]. Recent innovations permit visual assessment through a variety of approaches including late gadolinium enhancement, T1 and T2 mapping.

Increased attempts have been performed to use CMR for the guidance of invasive procedures[10]. CMR-imaging could enable major advantages when guiding in real-time cardiac electrophysiology (EP) procedures offering high-resolution anatomy, arrhythmia substrate, and ablation lesion visualization in the absence of ionizing radiation. Scar tissue characterization has a high correlation with the electroanatomic maps (EAM) obtained during the ablation procedures while CMR provides delimitation within the entire myocardial thickness compared with endocardial or epicardial surface electroanatomic maps alone[11,12]. Over the last decade, technologies and platforms for performing electrophysiology procedures in a CMR environment have been developed. To date, human reports on interventional CMR are limited to typical atrial flutter ablation as several limitations have not permitted a routine clinical use.

The aim of this article is to review the clinical implementation of real-time CMR-guided catheter ablation and to discuss the challenges and limitations in this early stage of this approach as well as the potential benefits and the future perspectives in the treatment of cardiac arrhythmias.

TECHNICAL ASPECTS

Performing procedures in a CMR environment and outside the conventional fluoroscopic laboratory posed technical, practical and safety concerns[13]. A number of limiting factors should be overcome as the development of magnetic resonance imaging (MRI) compatible ablation systems, the recording of high-quality electrograms despite significant electromagnetic interference and reliable methods for catheter visualization and lesion assessment.

Interventional CMR suite

To transform the pre-existing magnetic resonance imaging environment into an interventional cardiac MRI suite, all standard EP (recording system, displays and catheters) and anesthetic instruments should be replaced with nonferromagnetic alternatives to avoid potential risks and adverse incidences of both patient and health care personnel (Figure 1)[13]. Ferromagnetic instruments that cannot be replaced, as the non-MRI compatible radio frequency (RF) generators should therefore positioned outside the scanner room (Figure 1E)[13,14]. Communication between the operators and the radiologist at the MRI console may be facilitated by a compatible wireless communication system.

Additionally, modifications are probably required to the electrical installation to comply with safety guidelines that include to a touch-voltage less than 10 mV, isolation transformers for all wall power outlets, and a ‘protected earth' connection for every device[13].

Patient preparation, including femoral vein access and possible intubation, is performed in an adjacent zone outside the scanner room[13]. Importantly, a detailed procedural workflow should have been established for the safe performance of the procedure and recognition and management of potential complications. Notably, CMR enables an early recognition of complications as pericardial effusion.

Figure 1 Full transformation of the pre-existing magnetic resonance imaging environment into an interventional cardiac magnetic resonance imaging suite. A: Pre-existing diagnostic magnetic resonance imaging (MRI) scanner room;B: Pre-existing diagnostic MRI control room;C:Transformed interventional cardiac magnetic resonance (iCMR) suite;D: Transformed iCMR control room.E: The non-MR compatible RF generator including coolingpump positioned in the iCMR control room.EP,electrophysiological;iCMR,interventional cardiac MRI.Citation: Bijvoet GP,Holtackers RJ,Smink J,Lloyd T,van den Hombergh CLM,Debie LJBM,Wildberger JE,Vernooy K,Mihl C,Chaldoupi SM.Transforming a pre-existing MRI environment into an interventional cardiac MRI suite.J Cardiovasc Electrophysiol 2021;32: 2090-2096 [PMID: 34164862 DOI: 10.1111/jce.15128].Epub 2021 Jul 4.Copyright ? 2021 The Authors.Journal of Cardiovascular Electrophysiology published by Wiley Periodicals LLC.(Reproduced with permission)[13].

Catheter visualization

Catheter location in conventional EAM systems is visualized using magnetic-based sensing or impedance-based tracking and displayed on approximate geometries of cardiac chambers[15]. MR conditional diagnostic and ablation catheters are similar in appearance and function to conventional catheters, but include proprietary components to reduce MR-induced heating[16]. MR conditional catheters were initially created using a polyether block amide plastic body, copper wires and platinum electrodes[17] while the currently approved ablation catheter incorporates gold tip electrodes for energy delivery, recording of electrograms and pacing (Figure 2). During CMR-guided electrophysiology procedures, there are two methods of catheter visualization and intra-procedural guidance, active and passive catheter tracking.

Passive catheters are discerned by local susceptibility artifacts that are induced by para- or ferromagnetic materials placed near the tip of the catheter[18-20]. Optimized imaging protocols using a steady-state free precession imaging sequence at frame rates of 4-8 frames per second provide an adequate temporal resolution[18-20]. However, passive tracking permits a single plane real-time visualization[18,20]. Therefore, manipulation of the catheter requires a continuous manual selection of the appropriate image plane and a constant communication between the operator and the radiologist at the MR imaging console being time-consuming and prone to localization errors.

In contrast to passive tracking that is based on local susceptibility artifacts, active tracking uses integrated receiver lumenless solenoid micro-coils at the tip of the catheter to determine its location (Figure 2)[16,21,22]. These micro-coils act as point-source detectors of MR signals. Locating these coils is accomplished by acquiring the MR signal in the presence of applied magnetic field gradient and identifying the position of the most intense frequency-domain signal[16,22]. The main advantage of this technique is that enables automation of the tracking of the catheter for the localization of its position controlling the MRI scan plane in real-time (Supplementary material and Video). Moreover, high spatial resolution is provided using tracking rates up to 50 frames per second[16].

Electrogram fidelity in the MRI environment

Distortion of the electrograms within the magnetic field can make interpretation of both the surface electrocardiogram (ECG) and intra-cardiac electrograms (EGMs) unreliable[14,23,24]. Although hardware development over recent years has enabled ECG and EGMs acquisitions during MRI examination, interpretation and analysis of waveforms is limited. Signals are severely distorted during MRI scans due to the effects of magnetohydrodynamic (MHD) voltages, RF pulses and fast-switching gradient magnetic fields (Figure 3)[23,24].

Figure 2 9 Fr.bipole irrigated-tip ablation catheter with two magnetic resonance receive coils in the distal end for active magnetic resonance tracking. (Reproduced with permission from https://imricor.com).

Figure 3 Distortion of the electrograms within the magnetic field. A: Baseline noise of intra-cardiac electrograms recorded with the coronary sinus catheter (blue arrows indicate the maximally distorted signal);B: Gradient-induced artifacts that cause high frequency peaks during magnetic resonance scanning(orange arrows).ABLc: Ablation catheter;CATH2c: Coronary sinus catheter.

The MHD (or magnetofluid dynamic) effect is a result of the static magnetic field and the movement of charge carriers that induces a voltage across the blood vessels[23,24]. This induced voltage superimposes on the signals and appears primarily during the S-T phase of the cardiac cycle as it has been related to the blood ejection through the aortic arch which is perpendicular to the magnetic field and coincides with the occurrence of the T-wave of the ECG. The RF pulses (64-128 MHz) and the fast-switching gradients (33-50 mT/m, 20-100 T/m/s), which are required for MRI, both disturb the signals because of the voltages induced on the electrodes, wires and patient’s body.

Passing electrograms through several levels of filtering limited the noise on EGMs, in a previous study[25]. In detail, low-pass RF filters to reduce the 64-MHz RF signal from the MRI scanner were combined with a series of active filters (a low pass filter of 300 Hz, a high-pass filter of 30 Hz and a 60-Hz notch filter) to reduce gradient signal-induced noise[23]. For low-pass filtering, the highest-quality EP signals was obtained at the frequency of 120 Hz despite lower peak-to-peak signal amplitude[26]. Algorithms that overcome the limitations of state-of the-art methods and enable suppression of MR gradient artifacts and improve signal denoising quality have also been described including adaptive noise cancellation and non-linear Bayesian filtering[23,27].

Signal distortion should be taken into consideration especially for interpretation of EGMs after previous ablation attempts as double potentials (for typical flutter ablation) or abnormal potentials of low-amplitude as late potentials (for ventricular tachycardia ablation), although previous reports have presented detection of these ambiguous electrograms[21,28,29]. Moreover, as interpretation of surface ECG leads recording (that are usually connected to the recorder for rhythm monitoring and early detection of complications) may be impeded, additional monitoring should be used as pulse waveform.

Ablation lesion assessment

Lesions of radiofrequency catheter ablation can be visualized with CMR imaging[30,31]. The failure to create contiguous and durable transmural lesions has been held largely responsible for high recurrence rates[8,9]. Changes in tissue electrical impedance, electrode tissue contact and delivered power during conventional ablation techniques may not strongly correlated with the actual lesion size[32]. Electrical isolation may also be observed despite the presence of gaps in myocardial tissue after ablation that can be identified with MRI[6,7]. Thus, real-time lesion imaging is attractive as it could assess the ablation results and potentially provide a procedural endpoint.

Imaging with T2 mapping detects inflamed edematous tissue (Figures 4 and 5)[33]. However, T2-derived edema also corresponds to reversible lesions and is poorly correlated to long-term outcome as edema subsides progressively leading to electrical reconnections[34]. Several studies have reported on the extent of post-ablation T2-weighted signal that is greater in extent than delayed enhancement and overlaps with the areas of irreversible injury[30,31].

Figure 4 Imaging with T2 mapping detects inflamed edematous tissue. A and B: T2-weighted magnetic resonance images of the cavotricuspid isthmus in the RAO view before (A) and after (B) ablation showing edema in the ablation lesions,indicated by the yellow arrows.Citation: Bijvoet GP,Holtackers RJ,Nies HMJM,Mihl C,Chaldoupi SM.The role of interventional cardiac magnetic resonance (iCMR) in a typical atrial flutter ablation: The shortest path may not always be the fastest.Int J Cardiol Heart Vasc 2022;41: 101078.[PMID: 35800043 DOI: 10.1016/j.ijcha.2022.101078].Copyright ? 2022 The Authors.Published by Elsevier B.V.(Reproduced under the terms of the Creative Commons CC-BY license)[44].

Figure 5 Acute lesion after radiofrequency ablation of the right cavotricuspid isthmus. A: Balanced steady-state free precision sequence image of the cavotricuspid isthmus (CTI) immediately after ablation.White asterisk indicates pericardial effusion.White arrow indicates a prominent eustachian valve;B and C:T2-weighted images preablation (B) and postablation (C) showing signal intensity enhancement of the isthmus line (white arrows);D: Noncontrast enhanced T1-weighted image of the CTI depicts acute necrotic lesions as signal intensity loss (black arrow);E: Postcontrast early enhancement image shows hypoenhanced myocardium localized at the CTI,known as a microvascular obstruction as an acute ablation lesion sign;F: Phase-sensitive inversion recovery image depicts acute ablation lesion in terms of black,hypoenhanced myocardium;G: Postcontrast late gadolinium enhancement images led to partially enhanced radiofrequency ablation lesions (white edge) with black necrotic core (black arrow).Citation: Ulbrich S,Huo Y,Tomala J,Wagner M,Richter U,Pu L,Mayer J,Zedda A,Krafft AJ,Lindborg K,Piorkowski C,Gaspar T.Magnetic resonance imaging-guided conventional catheter ablation of isthmus-dependent atrial flutter using active catheter imaging.Heart Rhythm O2.2022;3: 553-559 [PMID: 36340492 DOI: 10.1016/j.hroo.2022.06.011].Copyright ? 2022 Heart Rhythm Society.Published by Elsevier Inc.(Reproduced under the terms of the Creative Commons CC-BY license)[40].

Late contrast-enhancement is used to detect lesion necrosis and T1-weighted imaging is thought to reflect true procedural success determining reversibility of lesions (Figure 5)[33]. However, the use of gadolinium-based techniques has significant disadvantages related to wash-in and wash-out kinetics of this contrast agent[35]. Late contrast-enhanced imaging demonstrates higher contrast-to-noise ratio between normal myocardium tissue and the lesion. However, edematous tissues as well as previous fibrotic areas can also become enhanced, impeding identification of gaps between lesions. Furthermore, estimation of complete delayed enhancement is time-consuming. Measurements during the initial phase of contrast void overestimate the transmural extent of lesions while regions of micro-vascular obstruction acquired approximately 26 min after contrast administration to accurately predict the chronic lesion volume in a previous report[30,36]. Finally, repeated injections of gadolinium-based agents in a single session is limited by clinical restrictions in their dosage, as well as effects on imaging from accumulated contrast agent.

In this way, there has been interest in the development of intrinsic (non-contrast)-based methods for ablation lesion assessment[30]. Imaging with non-contrast-enhanced T1-weighted pulse sequence with long inversion time was demonstrated to produce images of ablation lesions with readily visible contrast between the lesion core and normal myocardium and improved image quality for visualization of both lesions and anatomy (Figure 5)[30]. Importantly, unlike contrast-enhanced imaging, in which enhancement pattern changes over time, non-contrast based techniques can be repeated multiple times during a procedure.

MR thermometry is a technique for the monitoring of thermal treatments that utilizes the temperature dependent proton resonant frequency shift that occurs in water molecules[37]. MR thermometry has been shown to provide a direct assessment of ablation lesion extend in the myocardium[37]. The dimensions of the thermal lesions measured on thermal dose images were correlated with T1-weighted images acquired immediately after the ablation and at gross pathology in an animal study, although prediction of the lesion durability remains unclear[37].

Procedural workflows for real-time CMR-guided ablation of the cavotricuspid isthmus (CTI) have been proposed[21,38,39]. Pre-ablation balanced steady-state free precession three dimensions (3D) whole heart (bSSFP-3DWH) sequences without contrast provide the anatomy of cardiac cavities and large thoracic vessels with a selected acquisition window in ventricular diastole. Segmentation techniques may be used to derive the right atrial contour of this acquisition for integration into a navigation system. As a baseline for post-ablation imaging, T2- weighted images are also acquired prior to ablation. For guidance of the ablation catheter, the optimal planes are selected for the visualization of the CTI. A fourchamber view depicts the tricuspid valve and the distance to the interatrial septum while a long axis view the entire CTI length[38]. Views similar to the standard fluoroscopy views may also be used[39]. During active tracking, a dedicated sequence permits detection of the tip of the catheter and enables its manipulation along the CTI. A catheter is also placed into the coronary sinus for pacing maneuvers in order to verify isthmus block. For post-procedural imaging, the abovementioned methods have been described. Imaging with non-contrast-enhanced T1-weighted pulse sequence with long inversion time can be performed multiple times in case of identified gaps. Gadolinium may also be administered in the end of the procedure for lesion assessment.

CLINICAL IMPLEMENTATION OF CMR-GUIDED ABLATION

Over the last two decades, substantial progress has been achieved in real-time CMR-guided electrophysiology studies and ablation procedures. In Tables 1 and 2, reported animal and human studies are presented. Following successful experimental reports, several human studies have demonstrated that CMR-guided catheter ablation is feasible without fluoroscopic guidance and enables the concurrent visualization of the targeted anatomical structure and substrate as well as the ablation lesion. The first human reports, in order to establish a procedural workflow, have rationally focused on typical atrial flutter ablation taking into consideration the relatively simple access to the right atrium and CTI[18,21,28,38,40].

Cavotricuspid isthmus dependent atrial flutter ablation

Reports of conventional radiofrequency catheter ablation of CTI-dependent atrial flutter have revealed a high acute success rate up to 95% and a low recurrence rate[41,42]. Moreover CTI-ablation is a relatively safe procedure with low risk of complications. However, difficult cases of initial failed ablation and persistent CTI conduction are occasionally encountered. A complex isthmus anatomy has been considered as a cause of failure to achieve a complete ablation line[43]. Isthmus pouches that are frequently present, a prominent Eustachian ridge and large pectinate muscles may impede catheter stability and navigation to target sites leading to poor tissue contact and low RF energy delivery (Figures 5A and 6). CMR-guidance provides visualization of these anatomical obstacles and enables the optimal target ablation line selection taking also into consideration the length and thickness of the lateral, medial and septal CTI portion[44].

Despite initial difficulties due either to technical issues or to unachievable procedural endpoint and requirement of ablation completion under fluoroscopic guidance[18,21,28], the most recent and larger studies have shown that CMRguided CTI ablation represents a valid alternative to conventional ablation with an acute success rate of 93% to 100%[38,40]. Procedural times were comparable with fluoroscopy-guided treatment with similar results with regards to direct procedural success and short-term follow-up in a comparative study[38]. A steep learning curve was also demonstrated with a small number of procedures needed to achieve a level of competency and a meaningful gradual reduction of procedural duration[38].

Future perspectives

To date, no human studies have evaluated the use of real-time CMR-guided ablation apart from procedures performedfor typical atrial flutter. Broadening the application in the field of ventricular tachycardia (VT) would be of essential impact considering the widespread use of substrate-based strategies in VT ablation[45-47]. In the context of structural heart disease, surviving myocardium within areas of scar provide a substrate for reentry circuits[48]. Substrate-based ablation strategies have been shown to be as equally effective as activation mapping, which is often limited by haemodynamic instability and non-inducibility[49]. Even substrate ablation based only on the integration of preprocedural CMR has been shown to be feasible and efficient while recent studies have shown improved VT recurrencefree survival compared to standard ablation[50,51]. Importantly, the information obtained from the CMR shows the wall distribution of the scar within the entire myocardial thickness[11]. Therefore, implementation of real-time CMR-guidance could increase the efficacy of VT ablation contributing also to deciding on the optimal approach during the procedure (endocardial, epicardial or combined). The VISABL-VT, a prospective, single-arm, multi-center trial will investigate the safety and efficacy of RF ablation of ventricular tachycardia associated with ischemic cardiomyopathy in the CMR environment (ClinicalTrials.gov Identifier: NCT05543798).

Table 1 Published animal studies on real-time cardiac magnetic resonance guided ablation

Towards the application of CMR-guided ablation in the field of atrial fibrillation, an MRI-compatible cryoablation system has been developed by removing all ferromagnetic components (as the circular mapping catheter) of a commercially available cryoballoon, implementing a compatible steering mechanism for balloon deflection and placing the console for the system outside the scanner room[52]. A recent animal study has shown that the real-time CMR-guided cryoablation of the pulmonary veins is feasible and provides the ability to visualize the freeze-zone formation during the freeze cycle[53]. Pulmonary vein reconnection has been reported as the main cause of arrhythmia recurrence and thus, durable isolation has been a key determinant of clinical outcome in patients undergoing catheter ablation for atrial fibrillation[54]. It has also been demonstrated that electrical isolation may be observed due to local tissue architecture and/or anisotropy despite the presence of gaps in myocardial tissue and the recovery of conductivity can potentially lead to arrhythmia recurrence[6]. MRI has been shown to be able to identify gaps in ablation lines[54] while a report using realtime MR thermometry and thermal dosimetry demonstrated a strong correlation between thermal lesion and postablation T1-w images as well as with measurements at gross pathology[37]. Although further investigation is warranted, if MRI is able to assess lesion quality and durability, real-time CMR-guidance could improve effectiveness of catheter ablation.

Figure 7 Electroanatomic mapping systems compatible for use inside a magnetic resonance scanner have been developed. A: 3D electroanatomical activation mapping;B: Integration with real-time cardiac magnetic resonance imaging planes.(Reproduced with permission from https://imricor.com).

However, several limitations should be solved before an extended application of CMR-guided ablation including the requirement of compatible tools as defibrillators and trans-septal needles, the scarce clinical data and safety concerns posed by performing procedures with high complication rate outside the conventional environment. Custom-made actively tracked needles (incorporated a receiver antenna) have been described to enable transseptal puncture under realtime CMR guidance[55]. Recently, a deflectable intracardiac MRI-compatible guiding-sheath was developed to accelerate imaging during CMR-guided electrophysiological interventions while real-time CMR-guided pericardiocentesis using commercially available passive access titanium needles has also been described[56,57].

EAM systems compatible for use inside an MR scanner have been developed (Figure 7)[58,59]. The achievement of active tracking opened up all the strengths of fast EAM, including activation and voltage mapping[21]. Integration with real-time imaging of cardiac anatomy, arrhythmia substrate and ablation lesions permits a combination of electrophysiological and anatomic information. However, further innovation of these tools may be warranted in order to be comparable to the conventional mapping systems including signal fidelity and modules for correction of annotation.

CONCLUSION

Real-time CMR-guided ablation could offer a number of benefits including not only radiation-sparing procedures, but also evaluation of cardiac anatomy and substrate as well as assessment of ablation lesion formation, although further research is warranted for confirming the above-mentioned potential advantages. The feasibility of CMR-guided CTI ablation has already been demonstrated and potential expansion to other more complex arrhythmias, especially ventricular tachycardia and atrial fibrillation, would be of essential impact. However, several limitations need to be solved before application of CMR-guided ablation in a broad clinical setting, including signal fidelity and compatible tools, while innovations in EAM integration could enable the combination of the advantages of conventional electrophysiological and substrate-based approaches.

FOOTNOTES

Author contributions:Tampakis K, Andrikopoulos G and Kampanarou S wrote and revised the manuscript; Pastromas S, Sykiotis A, Pyrpiri C, Bousoula M, Rozakis D and Kourgiannidis G contributed to the collection of data; All authors have read and approve the final manuscript.

Conflict-of-interest statement:All the authors have no conflicts to disclose.

Open-Access:This article is an open-access article that was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution NonCommercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: https://creativecommons.org/Licenses/by-nc/4.0/

Country/Territory of origin:Greece

ORCID number:Konstantinos Tampakis 0000-0003-4609-5685.

S-Editor:Liu JH

L-Editor:A

P-Editor:Xu ZH