The central point: The DECAAF II randomized 843 patients with persistent atrial fibrillation at 44 centers in 13 countries to compare ablation guided by atrial fibrosis mapping (late-enhancement MRI) versus conventional pulmonary vein isolation. The result was negative: the fibrosis-guided strategy did not reduce arrhythmia recurrence. Recurrence was 43.0% in the fibrosis-guided group versus 46.1% in the conventional group (HR 0.95; 95% CI, 0.77 to 1.17; P = 0.63).
Clinical context: why map fibrosis?
Pulmonary vein isolation (PVI) is the foundation of atrial fibrillation ablation. However, in persistent AF, the recurrence rate after the procedure remains significant. The hypothesis that atrial fibrosis acts as an arrhythmogenic substrate led to a search for strategies beyond PVI, using late gadolinium-enhanced cardiac MRI (LGE-MRI) to identify and ablate fibrotic areas.
The original DECAAF study, published in 2014, had shown an association between the extent of atrial fibrosis detected by MRI and arrhythmia recurrence after ablation. DECAAF II was designed to test whether ablation targeting these fibrotic areas could improve clinical outcomes.
Main DECAAF II data
- 843 patients with persistent AF randomized at 44 centers in 13 countries
- Intervention group: pulmonary vein isolation + fibrosis-guided ablation (421 patients)
- Control group: conventional pulmonary vein isolation (422 patients)
- Primary outcome: atrial arrhythmia recurrence between 3 and 18 months
- Result: 43.0% vs 46.1% (HR 0.95; 95% CI, 0.77 to 1.17; P = 0.63)
- Complications: 2.2% in the fibrosis group vs 0% in the conventional group (P = 0.001)
Study design
DECAAF II was an international, multicenter randomized clinical trial. Eligible patients were between 18 and 85 years old, had persistent AF and a clinical indication for catheter ablation. Recruitment took place from July 2016 to January 2020.
Before the procedure, all patients underwent late gadolinium-enhanced cardiac MRI to quantify atrial fibrosis. Patients were then randomized in a 1:1 ratio.
Fibrosis-guided group
Operators performed pulmonary vein isolation and then ablated the areas of fibrosis identified by MRI. The success of fibrosis ablation was evaluated with a scoring system (fibrosis targeting score, from 1 to 5), where a score of 3 or higher indicated adequate ablation. In the analyzed population, 80.9% of patients reached this threshold.
Conventional group
Operators performed pulmonary vein isolation only, without access to the fibrosis information obtained by MRI.
Median participant age was 62 years. About 79% were men. Median AF history was 13 months. Median baseline atrial fibrosis was approximately 18% in both groups.
Main results
Primary outcome: no difference between strategies
Atrial arrhythmia recurrence (AF, atrial flutter or atrial tachycardia lasting more than 30 seconds) occurred in 175 of 407 patients (43.0%) in the fibrosis-guided group and 188 of 408 patients (46.1%) in the conventional group. The difference was not statistically significant (HR 0.95; 95% CI, 0.77 to 1.17; P = 0.63).
Secondary outcomes: also no difference
AF burden (the proportion of time in arrhythmia) was similar between groups: 0.04% in the fibrosis group versus 0.07% in the conventional group (P = 0.47). Quality of life assessed by the Toronto AF Severity Scale and SF-36 also showed no significant differences. The proportion of patients taking antiarrhythmics was similar in both groups.
Residual fibrosis: a relevant finding
One finding that deserves attention is the analysis of residual fibrosis after ablation. In the fibrosis-guided group, median residual fibrosis was 21.3% versus 25.1% in the conventional group. More importantly, each 10% increase in residual fibrosis was associated with a higher risk of recurrence (HR 1.21; 95% CI, 1.04 to 1.41). This suggests residual fibrosis has prognostic value, although the strategy of directly ablating it did not translate into measurable clinical benefit.
Safety: a warning signal
The safety profile favored the conventional group. In the fibrosis-guided group, 9 serious adverse events occurred (2.2%) versus zero in the conventional group (P = 0.001). Among complications in the fibrosis group, 6 strokes and 2 deaths were recorded. None of these events occurred in the group undergoing pulmonary vein isolation alone.
The authors note that ablation of fibrotic areas outside the pulmonary vein ostia involved additional catheter manipulation in the left atrium. This may have contributed to the increased thromboembolic complications observed.
Study limitations
Lack of operator blinding
Electrophysiologists knew which group each patient had been randomized to. This may have influenced intraoperative decisions in the conventional group, such as whether to apply extra lesions beyond pulmonary vein isolation.
Monitoring adherence
The protocol used smartphone ECG for arrhythmia monitoring. Patient adherence to monitoring declined during follow-up in both groups, which may have underestimated the true recurrence rate.
Relatively short follow-up
Follow-up was 12 to 18 months. In persistent AF, late recurrences are relatively common, and longer follow-up could reveal differences between the strategies.
Fibrosis heterogeneity
The authors acknowledge that MRI does not distinguish between different fibrosis types (interstitial, replacement, focal, diffuse). It is possible that only certain patterns are truly arrhythmogenic and that indiscriminate ablation of every area with late enhancement is not the ideal approach.
Implications for clinical practice
DECAAF II delivered a clear message: with current imaging and ablation tools, ablating MRI-identified atrial fibrosis does not improve outcomes compared with standard pulmonary vein isolation in persistent AF. In addition, the extra strategy was associated with more serious complications.
This does not mean atrial fibrosis is irrelevant. The association between residual fibrosis and arrhythmia recurrence confirms that the fibrotic substrate plays a role in maintaining AF. The challenge is to develop more precise methods to distinguish truly arrhythmogenic fibrosis from clinically inert fibrosis.
In practice, the study reinforces that complete and durable pulmonary vein isolation remains the cornerstone of AF ablation. Additional strategies extending ablation beyond the pulmonary vein ostia should be considered cautiously, weighing potential benefit against the added risk of complications.
My view as an electrophysiologist: DECAAF II is important because it rigorously tested a hypothesis that appeared promising. In daily practice, we encounter patients with persistent AF and extensive atrial fibrosis, and the temptation to “do more” during the procedure is real. This study reminds us that more ablation does not always mean a better outcome and patient safety must be the first criterion. The fact that residual fibrosis predicts recurrence is valuable, but we need better tools to act on it. Until then, well-executed pulmonary vein isolation remains the most balanced strategy between efficacy and safety in persistent AF.
