Cardiac MR: Closing the Gap in Redo Ablations

Despite improvements in technologies and techniques, some patients who undergo catheter ablations experience recurrence of their atrial fibrillation. Cardiac MR’s ability to detect ablation scarring may help guide redo procedures, but perhaps not just yet. 

Making progress

In less than a decade, MR has evolved from being an imaging modality for detecting ablation-related lesions in animal models to a bona fide player in human studies. It offers a noninvasive, high-resolution method for assessing lesions in the left atrium after pulmonary vein isolation. With catheter ablation’s success rates sometimes as high as 90-plus percent for paroxysmal atrial fibrillation yet lower for persistent atrial fibrillation, MR also may serve two roles: immediate feedback and guidance in repeat ablations.

“People want to see if immediately during a procedure, did we succeed and complete the ablation? That is one goal,” says Dana C. Peters, PhD, an assistant professor of diagnostic radiology at Yale School of Medicine in New Haven, Conn. “Then the other goal is after the ablation, when [atrial fibrillation] recurs, can we look and see which vein might be the target vein for further ablation.”

Peters and her colleagues showed in a study with 23 patients that high-spatial resolution delayed enhancement MR could detect atrial scarring after radiofrequency ablation to isolate pulmonary veins (Radiology 2007; 243[3]:690-695). In another pilot study, her team used 3D late gadolinium enhancement (LGE) cardiac MR to evaluate the relationship between scar volume and LGE distribution and outcomes in 35 patients (J Am Coll Cardiol Img 2009;2[3]:308-316). They found that the extent of scarring predicted recurrence of atrial fibrillation after ablations. On average, patients who experienced a recurrence had 2.6 noncircumferentially ablated veins compared with 1.8 for those who remained recurrence-free.

Repeat ablations, not mistakes

MR may help electrophysiologists refine their strategies when they need to perform a repeat ablation, proposes Saman Nazarian, MD, PhD, director of the Ventricular Arrhythmia Ablation Service at Johns Hopkins Heart and Vascular Institute in Baltimore. Like Peters, Nazarian has been at the forefront in exploring the feasibility of MR for electrophysiology purposes through novel animal and human studies. 

“What we are trying to accomplish is to identify reliably what had been done in the first procedure so in the second procedure we can skip the ablation of certain areas where we can see a good solid line of block from the first procedure,” he says. “Then, instead, we can devote our efforts to the areas where there isn’t enough ablation.”

Their initial experiences using MR to help in repeat ablations offered insights on MR’s potential and limitations (Heart Rhythm 2012;9:2003-2009). In 10 patients who underwent repeat pulmonary vein isolation, they preprocedurally performed delayed enhancement MR imaging of the left atrium and also obtained high-density voltage mapping and then co-registered the data.

The researchers found what they considered to be a reasonable correlation between the scar distribution and low-voltage areas on mapping. But they could not distinguish between pre-existing scar and ablation-related scar. They described the imaging resolution as limited, questioning the reliability of cardiac MR for identifying sites of early activation in the pulmonary veins.

Nazarian sees three camps for cardiac MR imaging and ablation: those who are convinced of its clinical utility already, those who think it may help but question its reliability and those who see no value in it. “I am convinced it would help, but I am not convinced based on the work I have done that we can [use it] reliably,” he says. “We need to work on the technique more. We need better resolution from MRI and we need better analysis techniques to make sense of the MRI before we can move this to a clinical [setting].”

Resolution & rewards

More recently, researchers have reported success in their efforts to improve MR’s reliability and utility. Using delayed enhancement cardiac MR, a European team showed they were able to target gaps and re-isolate pulmonary veins at a rate of 95.6 percent (J Am Coll Cardiol Img 2014;7:653-663). In an accompanying editorial, Nazarian and Johns Hopkins colleague Roy Beinart, MD, attributed the Europeans’ favorable results to differences in magnetic strength, the cardiac MR acquisition protocol and imaging post-processing. They suggested that discordant results on LGE and electrical mapping may be due to smaller, pre-existing scar and small gaps that are difficult to visualize.

Peters agrees MR has limitations but they likely are surmountable. “It is really a challenge to see the pulmonary vein and left atrial walls because they are 2 to 4 mm thick,” says Peters. “MRI doesn’t have the best resolution, but it is the best technique for this. It is almost at the edge of what we are able to see to really evaluate each vein.”

Cardiac and respiratory motion can compromise image quality, creating artifacts and making it difficult to see scar in the left atrium. Peters’ group is working on ways to improve MR’s image quality. “One method reduces the effect of heart rate variability in the MR images,” she says. “Sometimes people are in atrial fibrillation and the atrial fibrillation rhythm is very variable. We want to reduce the influence of the atrial fibrillation rhythm on images by basically taking into account the arrhythmia.”

Ultimately, it may be patients with persistent rather than paroxysmal atrial fibrillation who benefit the most from cardiac MR if the image quality and reliability can be maintained, Nazarian predicts. These patients often are more complicated and more likely to require repeat ablations. The potential incremental benefit of cardiac MR would impact these patients more.

“If the success rate of persistent atrial fibrillation is 50 percent right now, then any benefit you give them would be helpful,” he says. “The chances are higher that we will improve the care of patients with persistent atrial fibrillation.”