DECAAF: Emphasizing the importance of MRI in AF ablation

Atrial fibrillation ablation is a complex and challenging procedure. Appropriate patient selection is the most critical step to ensure safe and successful atrial fibrillation ablation procedure. The DECAAF study (Delayed-Enhancement MRI Determinant of Successful Radiofrequency Catheter Ablation of Atrial Fibrillation) showed that atrial tissue fibrosis, as estimated by delayed enhancement magnetic resonance imaging, was independently associated with recurrent arrhythmia post atrial fibrillation ablation. Magnetic resonance imaging also detected left atrial volume and shape. Integrating the data provided by magnetic resonance imaging into the pre-procedural planning is crucial.


BACKGROUND
Atrial fibrillation (AF) is the most common arrhythmia in clinical practice and is associated with increased morbidity, mortality, and additionally represents a financial burden in healthcare. Over the past decade, catheter ablation has become an important therapeutic modality for patients with symptomatic, drug refractory, AF. However, catheter ablation of AF is not risk-free, and potentially life threatening complications can arise, such as an atrio-oesophageal fistula, stroke, and cardiac tamponade 1 . The rates of these adverse events have been consistently dropping with the gained experience in ablation techniques over the last decade. Despite this growing experience, high rates of early and late AF recurrence remains the Achilles' heel of catheter ablation of AF 2,3 . Many studies have targeted the predictors of AF recurrence, such as patients' age and gender, left atrial size, type and duration of AF, left ventricular systolic dysfunction or heart failure, structural or valvular heart disease, and number of attempted ablation procedures 4,5 . Among these predictors, left atrial fibrosis has been established as an independent determinant for success of rhythm control strategies in AF. The DECAAF trial studied atrial tissue fibrosis estimation by delayed enhancement magnetic resonance imaging (DE-MRI) and its association with subsequent AF ablation outcome 6 .

DECAAF: DESIGN AND OUTCOMES
Delayed-Enhancement MRI Determinant of Successful Radiofrequency Catheter Ablation of Atrial Fibrillation (DECAAF) was a multicenter, prospective, observational cohort study of patients diagnosed with paroxysmal and persistent AF, and undergoing their first catheter ablation, conducted between August 2010 and August 2011 at 15 centers in the United States, Europe, and Australia. DE-MRI images were obtained up to 30 days before ablation. Fibrosis quantification was performed at a core laboratory blinded to the participating center, ablation approach, and procedure outcome. Cumulative incidence of recurrence was estimated by stage at days 325 and 475 after a 90-day blanking period, which is the standard time allowed for arrhythmias related to ablation-induced inflammation to subside. The risk of recurrence was estimated after adjusting for 10 demographic and clinical covariates.
In total, 260 patients with atrial fibrillation, including 65% with paroxysmal atrial fibrillation, were included in the study. All patients underwent a pre-ablation MRI up to 30 days before the procedure and 90 days following the ablation. Based on the degree of atrial remodeling/fibrosis, patients were classified into four groups: stage 1 (, 10% fibrosis), stage 2 (. 10% to ,20% fibrosis), stage 3 (. 20% to ,30% fibrosis), and stage 4 (.30% fibrosis) ( Figure 1). After adjustment for multiple variables, including age, gender, hypertension, comorbidities, type of atrial fibrillation, left atrial volume, and left ventricular ejection fraction, the extent of atrial disease was the only significant predictor of atrial-fibrillation recurrence. Each 1% increase in the extent of atrial fibrosis was associated with a significant 5.8% increased risk of recurrence post-ablation. For patients classified with stage 1 fibrosis, 84.6% were free from the arrhythmia at one year compared with just 31% of patients with stage 4 fibrosis. Overall, 64% and 54% of patients with stage 2 and 3 fibrosis were arrhythmia-free at one year.

DISCUSSION
There is a consistent body of evidence showing that left atrial remodelling has deterministic value in the development and progression of AF. It was proven since 1980's that the size of the left atrium correlates with AF incidence. Several studies have examined the relation between the echocardiographic left atrial dimensions and the development of AF. In the Framingham Heart Study, individuals in the highest quartile of left atrial diameter (.44 mm) had approximately 2.5 times the risk of developing AF compared with individuals in the lowest quartile of left atrial diameter (,35 mm) after adjusting for age and sex 7 . The Cardiovascular Health Study cohort observed similar results 8 . Furthermore, it was observed that patients with dilated left atria are also more likely to relapse after ablation of AF, making left atrial size a well established risk factor for AF recurrence after AF ablation 9 .
Recently, the concept of left atrial remodelling has evolved beyond simple left atrial size quantification. Single centre studies have proven efficacy of DE-MRI at the atrial level in evaluation atrial fibrosis as an arrhythmogenic substrate 10,11 . These findings were further confirmed by the current multicenter DECAAF trial.
Beyond scar detection by DE-MRI, a new non-invasive shape-based remodelling parameter, left atrial sphericity, which uses magnetic resonance angiography, was strongly and independently associated with AF recurrences post ablation 12 . In addition, the periatrial, periventricular and total pericardial fat volumes, detected by MRI, were found to be associated with poorer post AF ablation 13 .
Post-AF ablation, MRI can assess the degree left atrial reverse remodelling by detecting decrease in left atrial size 14 . Furthermore, in cases of AF recurrence, MRI can guide electrophysiologists to localise the reconnection sites at the ablation lines 15 (Figure 2).

DISCUSSION AND CRITIQUE
Although DE-MRI provides reliable staging of atrial disease, the inherent technical limitations of the imaging technique, such as the spatial resolution applied to the thin atrial wall, may limit its use on broader spectrum.
Moreover, no correlation was found between the extent of left atrial fibrosis and the duration or type of AF. Previous studies have demonstrated patients with paroxysmal AF have a large fibrosis, while other patients with persistent AF show only mild degrees of fibrosis. Similar results were observed in the DECAAF study, where the number of patients with paroxysmal AF was highest in the group with the most extensive fibrosis (79.2% paroxysmal AF in grade 4 fibrosis vs. 65.3% in grade 1). These findings do not support a causal relationship in which AF produces fibrosis, instead of being a consequence of a more complex left atrial fibrotic process 16 .
In addition, the method by which investigators detected AF recurrence is not clear. In fact, the diagnosis of AF recurrences based only on symptoms reported by patients is of no true clinical value. The absence of symptoms must not be simply interpreted as absence of AF. Moreover, many of the palpitations reported post AF ablation are due to supraventricular extra beats rather than AF episodes. Conventional Holter monitoring can only provide a snapshot documentation of the underlying heart rhythm for 24 hours up to 7 days, failing to detect arrhythmic episodes beyond this very limited time window. On the other hand, implantable continuous heart rhythm monitor devices (loop recorder) represent a promising option to detect the true AF burden post-ablation.
Further studies will be needed in order to implement the important results of DECAAF into a wider clinical context, for example, to evaluate the clinical outcomes of using high scar index as an exclusion criterion from AF ablation procedures. Other attractive research point is to integrate data on left atrial fibrosis detected by MRI with the electrophysiological properties of the left atrium assessed using voltage mapping and identification of complex fractionated electrograms.

WHAT HAVE WE LEARNT?
AF ablation is a challenging procedure. Appropriate patient selection is the most critical step to ensure safe and successful AF ablation procedure. Integrating data on left atrial anatomy, size, shape, and degree of fibrosis provided by MRI into pre-procedural planning is crucial (Figure 3). This is particularly imperative in centers still in their initial AF ablation experience. Yet, it is important to acknowledge other processes affecting the electrical, molecular, contractile and structural properties of the atria that are necessary for the initiation and maintenance of AF. Therefore, we need to recognize the data about left atrial fibrosis, and the whole concept of AF ablation, within a wider context in which AF is a complex, multi-factorial progressive disorder 17 .