Delayed Enhancement MR Imaging: A Potential ‘Game-Changer’ for Assessment and Treatment of Atrial Fibrillation

Start Page: 
28
End page: 
30
Author(s): 

Dr. Nassir Marrouche, Executive Director of the Comprehensive Arrhythmia Research & Management Center (CARMA), Director of Electrophysiology

Less than three years ago, my article in this publication1 (“MR Imaging: New Techniques for the Treatment of Atrial Fibrillation,” November 2008) noted that, “as advanced techniques for scar imaging become available, MRI will be an indispensable tool within the catheter lab itself.” Since then, research involving delayed enhancement magnetic resonance imaging (DE-MRI) has accelerated at an ever-quickening pace.

A number of studies conducted by the Comprehensive Arrhythmia Research and Management (CARMA) Center at the University of Utah have shown that MRI has tremendous potential to improve the diagnosis and treatment of atrial fibrillation (AF). MR angiograms of the left atrium (LA) help to improve anatomical maps, which are used in conjunction with other imaging modalities to determine optimal sites to target during ablation. In addition to the pre-procedure assessment, MRI has been used post-ablation to detect and quantify the extent of scarring in the LA. Scar formation and the location of that scar within the LA are indicators of procedural success and the prevention of AF recurrence. Without this technology, such assessments could only be made using invasive methods.

The most promising new development, however, is the application of DE-MRI as a pre-ablation investigative tool into the structural remodeling of LA wall tissue. When used with custom software tools and computer algorithms (Figure 1), DE-MRI not only calculates the extent of non-healthy LA tissue, but also predicts the success of first-time ablations as well as assists with the selection of optimal rate/rhythm control strategies. Given the costs of ablation procedures and heightened demands for replicable, measurable methods that ensure efficacy, such a prognostic tool has incalculable value. As Cardiovascular Business reported in its January 2011 cover story, “Delayed enhancement MRI could be a game changer.”2

Personalizing AF Management

An innovative DE-MRI visualization technique and analysis protocol developed at the CARMA Center personalizes AF management. By using custom-made, high-resolution DE-MRI sequences, CARMA researchers define areas of enhancement within the atrial wall, which correspond to intramyocardial fibrosis and/or ablation scar. This information is the basis for staging disease progression based on the amount of pre-ablation delayed enhancement (fibrosis) as a percentage of the volume of the LA wall. This clinical staging system includes four stages: Utah I (≤5% enhancement) Utah II (>5–20%), Utah III (>20–35%) and Utah IV (>35%) (Figure 3).  

This classification system is then used in concert with a comprehensive MRI-based AF management algorithm, which helps in triaging patients to AF ablation as well as planning a corresponding ablation strategy and future anticoagulation therapy. All potential ablation patients at CARMA are triaged using the Utah Classification System of Left Atrial Substrate Remodeling (Figure 2). We have found that approximately 5 to 12 percent of our patients are Utah IV, which makes them unsuitable candidates for ablation. Conversely, about 40 percent with persistent or permanent AF and minimal scarring (Utah I–III) are good candidates. The success rate for patients in Utah I is greater than 95 percent.

DE-MRI also allows for ablation lesion characterization. Immediate post-AF ablation acute edema, defined as enhancement on T2-weighted MRI, correlates with low voltage areas, but not with DE-MRI-defined scar. A study published in Circulation: Arrhythmia and Electrophysiology3 found that only 6.9 percent of 149 patients had complete and contiguous circumferential scarring of all four pulmonary vein antrum; DE-MRI accurately depicted the location and extent of scar lesions and identified the location of gaps in ablation lesions. Such knowledge can be helpful to electrophysiologists in planning repeat procedures.

Another CARMA study published last year in the Journal of Cardiovascular Electrophysiology found that the level of circumferential pulmonary vein scarring (Figure 4) in Utah II and overall left atrium wall scarring in Utah III predicted recurrence, while Utah IV had no recurrence predictors. The researchers concluded that when more atrial tissue is involved in creating the AF substrate, “going outside the pulmonary vein antral region to modify this substrate leads to improved ablation outcomes.”4

Screening for the Risk of Stroke

DE-MRI also has promise as a risk-assessment screening tool for those with AF at risk for stroke, the third leading cause of death in the US. Individuals with AF are five times more likely to suffer a stroke than the general population, with as much as 30 percent of the AF population being asymptomatic. Although the official rate for AF-related stroke is between 15 and 20 percent, neurologists (who are generally the first specialists to assess these patients) estimate this figure could be as high as 40 percent. Screening can identify asymptomatic patients before their disease progresses, and early intervention increases their chances of markedly improving their quality of life.

While further prospective studies are needed, preliminary research suggests that having more than 20 percent fibrosis in the LA may be predictive of heightened stroke risk. A study in the January issue of the Journal of the American College of Cardiology,5 for example, investigated whether there was an association between an AF patient’s LA fibrosis, which was detected using DE-MRI, and commonly used markers for the risk of stroke, specifically the CHADS2 index. It included 387 patients who were treated for AF at the University of Utah and the Klinikum Coburg (Coburg, Germany). The researchers concluded that LA fibrosis as determined through the use of DE-MRI is “associated with an increased risk of thromboembolism in AF patients. Clinician use of both a CHADS2 index and a quantified measure of atrial fibrosis has the potential to provide a more rigorous risk assessment and improve future risk stratification schemes.”5

The study also noted that clinicians can use this method in conjunction with the CHADS2 index for the selection of anticoagulation medications when treating AF patients. While the CHADS2 index is the most accepted risk stratification model and a valuable tool for predicting cerebrovascular events in high-risk patients, clinicians rely more heavily on clinical judgment when predicting thromboembolic risk in moderate risk patients, which includes a substantial portion of the AF population. The identification of novel, independent risk factors by DE-MRI may complement existing tools to help guide clinician judgment in selecting optimal anticoagulation therapeutic strategies, especially for moderate risk AF patients.

DECAAF Clinical Trials and HRS Studies

The most expansive research into DE-MRI image acquisition and its applications in the management of AF is the CARMA Center’s DE-MRI Determinant of Successful Radiofrequency Catheter Ablation of Atrial Fibrillation (DECAAF) multi-site international study, which includes several prestigious national and international leading academic medical centers (Figure 5). DECAAF is collecting data for the purpose of establishing the use of DE-MRI to stage AF (based on pre-ablation fibrosis percentages) and predict outcomes prior to AF ablation procedure. It has three specific aims:

  • Successfully acquire DE-MRI scans of the LA in AF patients and quantify the extent of structural remodeling or fibrosis.
  • Successfully acquire DE-MRI scans of the LA three months post-catheter ablation to quantify the extent of ablation-related scarring and describe its distribution.
  • Evaluate the relative contribution of clinical patient characteristics, pre-ablation fibrosis as well as post-ablation scarring, in determining the long-term success of catheter ablation in maintaining normal rhythm.


The Heart Rhythm Society’s 32nd Annual Scientific Sessions (May 4–7) in San Francisco will have 12 abstracts presenting the CARMA Center’s latest research on DE-MRI, the Utah Classification System and real-time MRI. Some of the highlights and results of these HRS presentations include:

  • Patients were classified into two groups with minimal scar (total LA scar <10%) and extensive scar (>20%), and classified based on LA fibrosis using the Utah Classification System. The study found that pre-existent LA tissue with fibrotic remodeling and dilation influenced the degree of radiofrequency-induced scar tissue that is formed following AF ablation.
  • DE-MRI was used to assess which clinical parameters were most influential in the LA remodeling process. Based on results from multivariate analysis, AF burden was shown to be the most important predictor for advanced staging classification (Stage III and IV). The study concluded that the duration of time spent in AF is a powerful predictor of AF structural remodeling and that LA dilation is associated with increased LA enhancement seen on DE-MRI.
  • Factors associated with the failure of multiple ablation procedures were evaluated using DE-MRI. The results showed that patients who failed repeat AF ablation had extensive pre-ablation LA structural remodeling. Although repeat procedures resulted in more LA scar, the increase in total LA scar and number of PVs with complete scarring did not influence repeat ablation success.
  • In a study using a porcine model, real-time MRI was used to identify gaps in ablation lesion sets and successfully target them acutely for ablation.
  • Exercise capacity was inversely associated with LA structural remodeling in patients with AF as detected by DE-MRI.

Limitations, Challenges and a Vision for the Future

A number of questions must be answered before electrophysiologists can adopt an MRI-based approach in their daily practices. A significant challenge for inexperienced MRI laboratories is scanning optimization for good quality atrial image acquisition. Another limitation at various institutions is accessibility to an integrated electrophysiology magnetic resonance imaging (EP-MRI) laboratory or cardiac MRI scanner.


Edsays: May 15.2011 at 08:32 am

Seems that the predictive ability of DE-MRI as it pertains to procedure success is highly dependent upon ablation operator experience and their ability to have consistent, reproducible results. What is the error rate in predicting success? I thought I read somewhere it was around +/- 15%.

Reply to this comment »

Post new comment

  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.

More information about formatting options

4 + 1 =
Solve this simple math problem and enter the result. E.g. for 1+3, enter 4.