Chapter 25. Atrial Fibrillation

Atrial fibrillation (AF) is the most common form of cardiac arrhythmia, so a review of the role imaging in AF is a natural topic to include in this book. Further motivation comes from the fact that the treatment of AF probably includes more different forms of imaging, often merged or combined in a variety of ways, than perhaps any other clinical intervention. A typical clinical electrophysiology lab for the treatment of AF usually contains no less than six and often more than eight individual monitors, each rendering some form of image-based information about the patient undergoing therapy. There is naturally great motivation to merge different images and different imaging modalities in the setting of AF, but this is also very challenging as a result of a host of factors, including the small size, extremely thin walls, large natural variation in atrial shape, and the fact that fibrillation is occurring so atrial shape is changing rapidly and irregularly. Thus, the use of multimodal imaging has recently become a very active and challenging area of image processing and analysis research and development, driven by an enormous clinical need to understand and treat a disease that affects approximately five million Americans alone, a number that is predicted to increase to almost 16 million by 2050.1

In this chapter, we attempt to provide an overview of the large variety of imaging modalities and their uses in the management and understanding of AF, with special emphasis on the most novel applications of magnetic resonance imaging (MRI) technology. To provide clinical and biomedical motivation, we outline the basics of the disease together with some contemporary hypotheses about its etiology and management. We then describe briefly the imaging modalities in common use in the management and research of AF, and then focus on the use of MRI for all phases of the management of patients with AF and indicate some of the major engineering challenges that can motivate further progress.

Clinical Profile of AF

AF is a growing problem in modern societies, with an enormous impact on both short-term quality of life and long-term survival.2 Approximately 0.5% of people age 50 to 59 years have AF, and 9% of people age 80 to 89 years have AF, and these prevalences are increasing.1,3 Although many people with the condition go untreated, AF is associated with an almost two-fold increase in the risk of mortality. AF patients experience a dramatically increased rate of stroke (from 1.5% for those aged 50 to 59 years to 23.5% for those between age 80 and 89),4 a risk that, by contrast, decreases with age in the normal population. Treatment of AF represents a significant health care burden, with the annual costs estimated at approximately seven billion US dollars.5

Restoring and maintaining normal sinus rhythm remains one of the major goals in treating patients with AF. One treatment modality is a combination of cardioversion and antiarrhythmic drugs6; however, only 40% to 60% of the AF population is maintained in regular rhythm 1 year after such treatment. The treatment itself may also have serious adverse effects7-9 and must usually continue for the lifetime of the patient. In contrast, maintaining sinus rhythm without the use of antiarrhythmic drugs seems to be associated with increased survival.10 The inadequacies of drug-based treatments for AF have long been the major motivation for finding a truly curative approach, that is, to maintain sinus rhythm and suppress AF.

Mechanisms of AF

The mechanisms underlying AF have been the topic of extensive research over many years, and there is consensus that the disease, like most cardiac arrhythmias, has two components, the tissue substrate and some initiating electrical events or triggers. Perhaps the most complete description of the substrate of AF comes from Wijffels et al, who first postulated that the longer the atria spend in the state of fibrillation, the more difficult it becomes to reverse the condition (ie, "Atrial fibrillation begets atrial fibrillation").11 Their conclusions were based largely on animal studies in which rapid pacing of the heart induced AF through a continuous process of electrical and then structural remodeling, a transition that is initially reversible but then becomes essentially permanent. Rapid pacing of the heart does not, of course, occur spontaneously, and the etiology of the disease in humans is thought to be closely linked to the gradual and inevitable increase in fibrosis in the atria that comes with age12,13 and that predisposes a heart to AF whether or not associated conditions such as heart failure are present.14 In animals subjected to the rapid pacing protocols developed to induce AF, treatment with a drug that suppresses the formation of fibrosis reduced the likelihood of developing AF compared with control animals,15 hence the clear link between fibrosis and the AF substrate. As we will discuss later, one application of imaging, especially MRI, is directed at identifying and quantifying the extent of fibrosis in the left atrium and thus identifying the progression of the disease substrate.

The role of triggers in AF is also motivation for novel imaging approaches. With electrical and electroanatomic mapping (ie, recording electrical activity from a number of known sites on a surface the heart), it is possible to identify the sites of triggers and thus also localize causes of induction of AF. Leaders in mapping triggers of AF are Haissaguerre and coworkers,16-19 who identified trigger sites both in the atria and especially within the pulmonary veins of the left atrium. Once identified and localized, it is possible to electrically isolate these triggers, which was the unknown consequence of an earlier surgical technique for AF management, known as the Cox maze procedure, first performed in 1988.20 This operation isolates not only the pulmonary veins but also different regions of the atria by creating a "maze," a tortuous path for electrical conduction in the atria, reducing the ability of triggers to interact with a substrate that could sustain arrhythmias. Modern, catheter-based ablation approaches seek to achieve similar goals by applying very focused energy to the endocardial surface of the atria to isolate the triggers known to exist in the pulmonary veins17 and create the same maze of broken conduction in the left atrium.16,21,22 The electroanatomic mapping approaches necessary to guide such interventions are, however, invasive and often time consuming, so there remains a pressing need to develop noninvasive imaging approaches to localize trigger sites. Current research in AF management seeks to develop imaging based on MRI,23-25 ultrasound,26 and computed x-ray tomography27 to visualize the ablation lesions and thus direct the intervention.

Imaging and AF

We outline in subsequent sections the use of a broad range of imaging modalities in AF management and then focus on a comprehensive approach to the management of AF using MRI for all phases of evaluation and intervention. The most novel approaches to imaging in AF are in the areas of merging multiple modalities and in the rapid expansion of the use of MRI. The rationale for this growth is that, unlike other modalities, MRI is naturally suited to detect changes in soft tissue characteristics and hence capable of revealing the progress of substrate in AF and in visualizing the creation of lesions during ablation. These capabilities, combined with the ability to reveal atrial anatomy at high resolutions, make MRI the natural adjunct to all phases of management of AF patients.

The use of imaging is ubiquitous in clinical electrophysiology, especially in interventions that require remote access to the heart by means of catheters. On one hand, imaging is required to guide the catheter, and on the other hand, the catheter itself often captures and conveys functional and diagnostic information that must be integrated into the procedure. Figure 25–1 summarizes the most common modes of displaying information in a typical electrophysiology study, and we describe here briefly these modalities. The focus of subsequent, more detailed discussion will be MRI, the modality that is now the topic of extensive research and development.

Fluoroscopy

Fluoroscopy is an x-ray–based modality that has been a mainstay of cardiac catheterization procedures from their inception. Like all x-ray–based imaging, fluoroscopy can reveal dense materials like bone and metallic objects (eg, electrodes, devices, catheters) but, without contrast agents, is not capable of visualizing soft tissue or blood. In catheter ablation, fluoroscopy serves primarily to guide catheter navigation and to direct the transseptal puncture of the atrial septum that is necessary to access the left atrium. With contrast agent injection, it is also possible to visualize vessels and cardiac chambers using fluoroscopy. Significant strengths of fluoroscopy include the ability to perform real-time imaging at frame rates of tens per second, the simplicity provided by very evolved technology, and ready availability. Acknowledged weaknesses include the poor soft tissue contrast and the cumulative exposure to ionizing radiation. More fundamentally, fluoroscopy is a two-dimensional modality so that revealing three-dimensional cardiac shape is challenging.

Modern fluoroscopy systems seek to address these limitations by providing multiple, orthogonally oriented cameras (biplane fluoroscopy) and rotational angiography systems for full, three-dimensional reconstructions of venous and cardiac shape during catheterization procedures,28,29 all at the lowest possible field strengths. The anatomic models obtained by rotational angiography systems approach the resolution and accuracy of computed tomography (CT) and MRI images. Furthermore, because imaging occurs intraprocedurally, the resulting models can be better aligned to the coordinate system in which the procedure occurs and are less vulnerable to intravascular volume changes and shifts in body position and shape that may limit the accuracy of remotely acquired MRI and CT images.30,31 Acquisition of geometric information in the same reference frame and at the same time of procedures significantly aids the integration of imaging modalities by obviating the need for registration.32

Electroanatomic Mapping and Imaging

Electroanatomic mapping (EAM) is an essential component of cardiac ablation procedures that has been used widely since the mid-1990s.33-38 All such systems produce a patient-specific geometric model of the endocardium together with electrograms at numerous sites on that surface. There are two competing EAM technologies, CARTO from Biosense Webster (Diamond Bar, CA)39 and EnSite from St Jude Medical (St. Paul, MN),33 which differ in the manner by which they generate the electrical signals on the endocardium. CARTO measures potentials directly by touching a manually steered catheter sequentially to the heart surface and can thus be used on both the endocardial and epicardial surfaces. Its major weakness is the time required to sample enough points to create true maps of electrical activity, which can present challenges when the underlying arrhythmias are unstable and poorly tolerated by the patient. The EnSite technology differs in that it is based on simultaneous recording from an inflatable catheter containing 80 electrodes that is placed inside the chamber of interest. The system solves the resulting bioelectric field inverse problem in terms of endocardial potentials from a single heartbeat, reducing the burden on patients with unstable arrhythmias.

More recently, clinicians have adopted EAM for evaluation and guidance during ablation of AF,34-36 but here, the goal is often simpler—to measure only the amplitude of electrical activity in the posterior wall of the left atrium and thus evaluate the success of ablation. In the setting of MRI-guided AF ablation, the role of EAM will likely change as the MRI itself can provide direct evidence of the extent of ablation lesion so that mapping of voltage will become of secondary importance.

Echocardiography/Ultrasound

Various modalities of ultrasound-based echocardiography are used in the management of AF. Transthoracic echocardiography (TTE), transesophageal echocardiography (TEE), and intracardiac echocardiography (ICE) are routinely used before, during, and after catheter ablation of AF, always with the goal of providing detailed and fine-scale anatomic information. We will briefly describe the context in which each of these modalities is commonly used.

Transthoracic Echocardiography

TTE is routinely used for screening and evaluation purposes in the management of AF, typically to screen for underlying heart disease, including heart failure, valvular heart disease, and left ventricular hypertrophy.40 Additionally, TTE can be used to assess left atrium size and anatomy.41 Postablation TTE can be used to detect pericardial effusion and to evaluate left atrial function and size.42 Although other imaging modalities outperform TTE in these tasks, TTE remains an effective, readily available, noninvasive, and relatively inexpensive modality.

Transesophageal Echocardiography

TEE has shown high sensitivity and specificity for the detection of left atrial thrombus before ablation treatment.43 It is also useful for assessment of the location and number of pulmonary veins when CT and MRI are not feasible. Due to patient discomfort (the probe must be placed in the esophagus at the level of the heart) and need for airway management, TEE has not traditionally been used intraprocedurally.42

Intracardiac Echocardiography

ICE has become a standard imaging utility in most modern electrophysiology laboratories because of its high spatial and temporal resolution, achieved in part from the immediate proximity of the sensor and the heart. ICE is capable of visualizing the anatomy of the left atrium, including the pulmonary veins and appendage, as well as other local anatomy including the aorta, mitral valve, and esophagus. In AF ablation procedures, the ultrasound catheter is navigated intravenously to the right atrium and is used to guide transseptal punctures, navigate ablation catheters, confirm electrode-tissue contact, and titrate energy delivery.44-47 Additionally, ICE plays a critical role in the prevention and detection of complications by monitoring the formation of thrombus or coagulum, pericardial effusion and tamponade, and flow acceleration indicative of pulmonary vein stenosis.48-51 Limitations of ICE include the requirement for additional intravenous access and confinement to two-dimensional imaging.

Anatomic MRI/Multislice CT

Multislice CT (MSCT) and MRI-based angiography are routinely performed before ablation to define left atrial and pulmonary vein anatomy and size. Models of the relevant cardiac anatomy, generated from these images, are created to help guide intraprocedural navigation and tissue targeting.52-54 Trade-offs exist between the selection of MSCT or MRI for angiography. MSCT-based angiography is faster and has higher spatial resolution, whereas MRI does not require exposure to ionizing radiation. Most patients with pacemakers or implantable cardiac defibrillators are also ineligible for MRI. High-fidelity representations of the anatomy can be generated from both modalities, and consequently, both are considered acceptable for this purpose. Both modalities are also used after ablation to identify complications such as pulmonary vein stenosis, atrioesophageal fistula, and reverse remodeling (ie, decrease in atrial volume).42,55-59 MRI and MSCT have also been successfully used to identify surrounding structures that may be at risk of collateral injury during AF ablation, including coronary vessels and the esophagus.60-62

Merging of Modalities

Clearly, no imaging modality is a panacea for all of the requirements inherent in the assessment and treatment of a disease with such diverse and complex imaging needs as AF. Often a merging of modalities is necessary or at least desirable to achieve the necessary coverage of anatomic and functional information to manage the disease. The challenges presented by merging imaging modalities include the need to align or register images acquired in different coordinate systems and different resolutions and then to present them to the operator in a way that is flexible and intuitive enough to be useful.

The centerpiece of contemporary integrated image merging systems tends to be the EAM system, which includes the necessary merging, registration, and visualization hardware and software. The goal of such systems is almost always to use a previously acquired angiography (by MRI or MSCT) to provide a geometric substrate for the subsequent EAM of the heart.63-69 This registration step usually relies on operator identification of landmarks common to both the angiography and the EAM to rigidly align them in the coordinate system of the EAM system. Further refinements of the alignment are then updated as more points are sampled for the EAM. Much of the mismatch that remains can be attributed to the differences in the MSCT and MRI data acquired sometimes days before the procedure. Other sources of error in the match of MRI and MSCT models to intraprocedure anatomy include respiration, patient movement, and changes in cardiac rhythm.70,71 Real-time integration of ICE imaging with the EAM system has recently emerged as a means to allow intraprocedural generation and updating of anatomic models, further improving navigational accuracy.72,73

Cardiac MRI has become the gold standard for imaging and analysis of numerous cardiac conditions. Generally, MRI is limited by comparatively slow image acquisition and reconstruction times, low resolution, susceptibility to noise, and magnetic field incompatibility of some patients. However, the two primary benefits of MRI are soft tissue contrast and absence of ionizing radiation. These two strengths come to bear significantly in the arena of AF management. First, AF is known to influence cardiac structural properties in a process known as remodeling. Second, the stated goal of AF ablation is to modify, isolate, or abolish arrhythmogenic tissues. In both cases, the soft tissue contrast available in MRI provides insight for physicians into the entrenchment of AF and success of scar formation, respectively. Finally, modern ablation procedures still rely heavily on fluoroscopy for procedural guidance. The introduction of AF ablation procedures into the MRI environment opens the door for the departure of ionizing radiation from the management of AF.

Late gadolinium-enhanced (LGE) MRI is used to evaluate alterations in tissue structure associated with numerous cardiomyopathies.74,75 To acquire LGE images, a dose of chelated gadolinium contrast agent is administered intravenously as would be done for standard magnetic resonance angiography. Following the injection, the gadolinium is allowed time to wash clear of normal myocardium, and an inversion-recovery prepared gradient echo pulse sequence is applied to detect regions of tissue where the contrast agent remains sequestered. Any region in which perfusion has decreased or extracellular space has increased will appear bright in LGE images due to enhanced concentrations of gadolinium relative to surrounding tissues.76

Evaluation of Postablation Scar Formation

The first reported use of LGE-MRI of atrial tissue to assess ablation lesions came from Peters et al.77 In this prospective study, contrast enhancement was found in the left atrium and pulmonary vein ostia of all patients who had undergone radiofrequency (RF) ablation of AF 1 to 3 months previously. These findings were supported by McGann et al,78 who quantified the extent of enhancement observed in the left atrial wall 3 months after ablation using the methods outlined in Fig. 25–2 and compared extent of scar to procedural outcomes. In this study, patients who experienced a recurrence of AF were found to have less enhancement (12.4 ± 5.7%) compared with patients who did not experience recurrence (19.3 ± 6.7%, P = .004).78 Subsequent studies have expanded on these initial findings to show that lesion remodeling stabilizes by 3 months after ablation and that the extent and continuity of lesions encompassing the pulmonary veins play an important role in preventing recurrences.79-81

The ability to noninvasively assess the lesion sets created in AF ablation procedures can provide valuable feedback to electrophysiologists searching for the optimal ablation strategy. Although freedom from AF after a single intervention will remain the goal for procedural success, LGE-MRI can help explain how and why a particular lesion set succeeds or fails at terminating AF and provide direction in subsequent ablation procedures.

Evaluation of AF Substrate

As previously noted, AF is associated with structural remodeling of the left atrium. Motivated by the success of LGE-MRI in identifying structural heart disease and, in particular, fibrosis, Oakes et al82 analyzed LGE-MRI scans from a cohort of 81 AF patients and six normal volunteers to explore the relationship between contrast enhancement and AF structural remodeling. This study revealed a positive correlation between low-voltage tissue regions in EAMs (bipolar voltage amplitude ≤0.5 mV) and left atrial wall enhancement (r2 = 0.61, P < .05). Furthermore, patients with mild (<15%, n = 43), moderate (15%-35%, n = 30), and extensive (>35%, n = 8) amounts of left atrial wall enhancement were found to have significantly different rates of AF recurrence at a mean follow-up of 9.6 months (14%, 43%, and 75%, respectively). These findings suggest that the degree of left atrial wall enhancement, which is assumed to reflect extent of fibrosis in the atrial tissue, is a predictor of failure for ablation. Based in part on these results, a staging system for determining the amount of enhancement has been proposed. Figure 25–3 shows the Utah staging system, with examples of LGE-MRI scans from each of the four stages. Under this system, patients with Utah stage III or IV enhancement are not considered to be ideal candidates for ablation therapy.

The utility of the Utah AF stages is currently under extensive evaluation, both at our institution and through a multicenter clinical study involving major AF centers from around the world.

As outlined in the Introduction, catheter-based ablation of the left atrium represents the most common intervention to cure or at least suppress the symptoms of AF. To carry out ablation requires considerable imaging support in order to identify anatomy, evaluate substrate, and determine success of the intervention. Conventional approaches to AF ablation make use of fluoroscopy, CT, intracardiac ultrasound, and EAM. Not only is MRI used as a preprocedural method to generate anatomic images, but also its broader use represents the leading edge of research in real-time imaging to support the guidance of catheters and the evaluation of lesion formation in a three-dimensional form. The need to combine anatomic and functional information continues to drive the development of novel merging and registration approaches.34

Catheter Ablation of AF

The past decades have seen significant progress in understanding the underlying mechanisms of AF that sustain its persistence,11,17 and this knowledge has led to the treatment paradigm of AF ablation, which is the targeted destruction of tissue predominantly in the left atrium in order to isolate electrical triggers and reduce the ability of the atrium to sustain rapid activation. AF ablation, typically based on RF energy delivery through a venous catheter, has already produced encouraging results and is the topic of innumerable research reports, but it has yet to reach its full potential. Despite the fact that ablation, when successful, allows the patient to discontinue the use of antiarrhythmics and anticoagulants, the success rate of ablation in maintaining regular sinus rhythm without the use of such medications is still only 60% to 80%.83 Moreover, the penetration of ablation, although difficult to measure with accuracy, appears to lie well below the need; there are fewer ablations carried out each year than there are new cases of AF. In an effort to increase the penetration of this potentially curative approach, there have been many modifications to the ablation procedure aimed at improving outcome and hence promoting the adoption of the ablation approach.16,34,72,84-89 Despite such progress, daunting technical challenges to carrying out successful ablation remain, and many of these are related to imaging.

Currently, AF ablation is performed using catheters that can be visualized under fluoroscopy and/or projected onto a three-dimensional virtual shell acquired through EAM during the procedure.34 There are multiple challenges associated with these approaches. First, it is impossible for the operator to visualize the catheter tip/tissue interface; hence, delivery of RF energy is based on guidance from the morphology of local electrogram or by using the virtual shell from EAM to assure that the catheter tip is in contact with the atrial wall. However, both of these approaches have known errors that can exceed 1 cm,90 leading to frequent delivery of inappropriate lesions that may only partially damage the atrial tissue, promoting tissue recovery and hence reoccurrence of the arrhythmia. Moreover, this lack of visualization of the catheter tip can result in localized heating of blood, thus leading to char formation, a major cause of embolic stroke during the ablation procedure.91 Defining a technology or a system that would allow accurate visualization of the catheter tip/tissue interface would overcome this major problem for the operator. Another major challenge of the ablation procedure is the lack of an imaging modality that allows immediate assessment of tissue damage as the RF energy is applied. MRI is the most obvious and perhaps only imaging system that could overcome this problem.

Although MRI has the inherent capability of visualizing soft tissue and thus providing both anatomic and functional guidance for RF ablation of AF, there are challenges to creating a viable MRI-based approach. First, it is necessary to develop an MRI-compatible catheter and associated software that allows visualization of the catheter during navigation and energy delivery within the atrial chamber. This catheter and software must be part of a system that tightly integrates the diverse instrumentation required to complete a clinical atrial ablation procedure. The system must exploit the benefits of soft tissue contrast unique to MRI (near real-time visualization of myocardial interfaces and ablation lesions) while providing a smooth workflow for the physician and technicians. Recent reports showing progress toward these ends by our and other groups23-25 suggest that a full MRI-guided AF ablation procedure in humans, although still very challenging, is likely to represent the future of this methodology.

MRI-Compatible Catheters

The development of an MRI-guided system for AF ablation is completely novel in terms of the devices and support systems that are required to create a working system. MRI-compatible catheters have just begun to appear in the literature23,33,5292 but are still prototypes and have not been used in any human ablation studies. Similarly, the real-time MRI guidance systems required to place the catheters in the appropriate locations are in their infancy, with only sparse reports of placing an MRI-compatible catheter in the human heart under MRI guidance.52 MRI-compatible versions of most of the other elements of the contemporary AF ablation instruments—the lasso catheter, the coronary sinus catheter, and the needle required to carry out transseptal punctures—are also only just under initial development and have yet to receive approval for use in humans.

We have participated in the development of catheters that are MRI compatible, steerable in a way similar to standard clinical catheters, and capable of both delivering RF energy and recording endocardial electrograms.25 It is possible to track the location of the catheters and to display their position superimposed on the real-time MRI images and with a geometric shell model of the atria and great vessels that we create from volumetric MRI scans recorded in the early phases of the procedure. Figure 25–4 shows an example of MRI images recorded during a real-time ablation procedure in which the catheter is visible superimposed on the orthogonal MRI images. Also visible in the image is a polygonal surface or shell of the right atrium and inferior and superior vena cava, created by segmenting a previously acquired high-resolution scan of the animal's atrial anatomy.

Visualization of Imaging Results

Scientific visualization is an essential step in using imaging data, and the unique and challenging needs of AF ablation continue to drive new approaches. For example, EAM requires the integration of spatial information describing the shape of the endocardial surface with time signals, that is, electrograms recorded from that surface, and parameters extracted from the electrograms. In addition, there is volumetric information from MRI and CT that can be visualized as a sequence of two-dimensional images but is much richer when rendered in a three-dimensional form. Naturally, there is a need to merge these two (and other) forms of anatomic and functional information, and novel visualization technology continues to improve such merging in a setting of interactive manipulation and rendering.34

We have developed techniques that combine not only visualization of volume- and surface-based approaches but also projection of the information from the volume to the surface. Figure 25–5 shows an example of such a visualization in which Fig. 25–5A shows an EAM from an animal experiment showing three clusters of lesions performed under fluoroscopy guidance; the red dots in panel A show the lesion sites. Figure 25–5B shows the LGE rendering of the same heart, performed in the MRI scanner directly after the ablation procedure. We then used segmentation of the volume images to define the endocardial surface and projected the information from the MRI scan onto that surface, shown in Fig. 25–5C. Confirmation of the actual lesion locations is evident from dissection documented in Fig. 25–5D.

The motivation of such projection approaches is to present information from multiple modalities (in this case, EAM of electrogram amplitude and MRI tissue changes) in a common reference frame (in this case, the endocardial surface). Such merging of information provides for quantitative analysis and comparisons and also a means of conveying information in a form that is familiar to the clinicians (in this case, electrophysiologists who are highly conversant in the conventions of endocardial and epicardial mapping).

Real-Time Detection of Lesion Formation

The most significant advantage of MRI-guided ablation is its potential to obtain rapid feedback on tissue changes during the ablation procedure—to watch the lesions form. There is no viable modality at this time that can determine the effectiveness of ablation; even electrical mapping approaches only measure depressed electrical activity in the endocardium that may return within weeks of the ablation and cause a recurrence of AF. MRI, on the other hand, has the potential to visualize changes in tissue structure related to permanent cell damage following the application of energy and thus establish the presence and depth in the atrial wall of terminally destructive lesions. Visualization of lesion formation and extent would improve the effectiveness and the safety of RF ablation procedures.

To visualize lesion formation with MRI requires acquisition of high-quality images in rapid sequence in order to achieve adequate spatial and temporal resolution. As with all imaging modalities, there is a trade-off in MRI between the time needed to acquire the image and the quality of that image. Furthermore, real-time imaging of the heart is driven both by the need to capture information rapidly enough to avoid blurring due to cardiac and respiratory motion and the desire to optimize image quality to see small changes within structures that are only a few millimeters thick. Another challenge specific to ablation is the need to see changes quickly enough to allow the operator to control the time and the energy dose in order to create lesions that are deep enough, but not so deep as to degrade the structural integrity of the heart wall.

In animal studies within our group, we have achieved image refresh rates of up to 5.5 frames per second based on customized MRI scan sequences and have been able to visualize lesion formation within 10 to 15 seconds of onset of RF energy.25,93 Figure 25–6 shows just one example of such a case, in which catheter placement is documented in Fig. 25–6A, followed by a sequence of images (using a different MRI scan sequence) that reveal the formation of the lesion in Fig. 25–6B to 25–6F. Figure 25–6G shows a postmortem image in the same plane, and Fig. 25–6H contains a photographic record of the lesion seen immediately after the experiment. To our knowledge, this is the first report of visualizing lesion formation as it occurs in any tissues of the heart. We have also compared lesion sizes measured from MRI imaging with those determined through postmortem dissection and shown excellent agreement.25

Imaging has always been an essential component of the management of all forms of cardiac arrhythmias, and its use will continue to expand in pace with improvements in the imaging acquisition technology, the image processing and analysis, and the integrating software that can efficiently support the clinical workflow. In the setting of AF, the use of MRI is making particular advances as its utility in all phases of the disease becomes evident. Preablation imaging provides a means to stage patients and determine their best treatment options; the emergence of the Utah AF staging system suggests a very specific means by which image analysis and quantification can indicate disease status and risk. Similar techniques provide a means of noninvasively determining the outcome of AF ablation by mapping the formation of scar to both evaluate interventional success and to guide subsequent interventions, should they be necessary. The Comprehensive Arrhythmia Research and Management Center has carried out over 600 scans on over 250 patients to date and is now collaborating with similar laboratories around the world in multicenter trials of these MRI-based approaches. The results of these studies could completely transform the way that AF is treated when it arises and even enable preventative measures that are simply impossible without a means of tracking the tissue changes that preface the onset of electrical symptoms.

The potential for imaging, especially MRI in the treatment of patients with AF is equally exciting and bright. We have carried out over 30 animals studies in developing the prototype MRI-guided navigation system and are focused on developing and testing such a system for use in humans. We have created lesions both under fluoroscopy and real-time MRI guidance in the atria and ventricles of anesthetized dogs and swine and then carried out detailed imaging both of the entire animal thorax and of the excised preserved heart. In the process, we have made advances in MRI-compatible catheter design, exploitation of novel sensing coils, incorporation of tracking coils into catheter housings, improvement in pulse sequence design for rapid acquisition, integrated interactive display of images and devices, and image processing and analysis tools for postprocedure evaluation of results. Other groups have made similar progress, and there is little doubt that the first human studies are immanent.

A major initial goal of these studies was to ensure that it is indeed possible to visualize lesions soon after ablation, despite the thin atrial walls and small extent of RF lesions. Figure 25–6 shows an example of such a result in which we sampled from the same slice before, during, and repeatedly after application of the RF energy.93

The authors are deeply indebted to the contributions of the other members of Comprehensive Arrhythmia Research and MAnagement (CARMA) Center at the University of Utah for this material (www.carmacenter.org). The mission of CARMA is to provide worldwide pioneering leadership in advancing clinical treatments and research for cardiac arrhythmias, especially atrial fibrillation. We also thank the Cardiovascular Research and Training Institute (CVRTI) for their support in experiments of MRI-guided ablation.

Support for this research comes from the NIH NCRR Center for Integrative Biomedical Computing (www.sci.utah.edu/cibc), NIH NCRR Grant No. 5P41-RR012553-10 at the Scientific Computing and Imaging (SCI) Institute and from research grants from Surgivision Inc. and Siemens Healthcare.