We retrospectively reviewed all patients referred to 3 centers (West China Hospital, Beijing Anzhen Hospital, and Beijing Chaoyang Hospital) for catheter ablation of persistent AF between 2020 and 2022. Reports from transthoracic echocardiography performed within one week of cardiac catheterization were screened with the following inclusion criteria: (1) moderate or severe secondary MR; (2) LV ejection fraction (LVEF) $>$50%; (3) LV end-diastolic diameter 55 mm; (4) systolic LA anteroposterior diameter $>$35 mm; (5) normal leaflet anatomy and motion. Patients with intrinsic mitral abnormalities, including rheumatic, degenerative, congenital mitral valve disease, or extensive calcification involving mitral leaflets and annulus, were excluded from this study [8]. Other exclusion criteria were LV regional wall motion abnormality (assessed based on wall thickening and endocardial motion of myocardial segment in echocardiography) [9], myocardial infarction, or cardiac surgery within 3 months. Demographic and clinical information were prospectively obtained in all patients. All participants provided informed consent before the ablation procedure alongside authorization to use their data for research purposes. The study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Ethics Review Committees at West China Hospital, Beijing Anzhen Hospital, and Beijing Chaoyang Hospital.

Transthoracic echocardiography was performed according to published guidelines [8, 9, 10]. LA anteroposterior systolic diameter was measured in the parasternal long-axis view, and right atrial (RA) transverse systolic diameter was measured in the apical four-chamber view. LVEF was calculated using the modified Simpson’s method. The severity of MR and tricuspid regurgitation (TR) was defined using a multiparametric approach, including assessing vena contracta width, effective regurgitant orifice area, and regurgitation volume. MR and TR were graded as none, mild, moderate, or severe.

Patients were on oral anticoagulation for at least 1 month before the procedure, and transesophageal echocardiography was performed to exclude LA thrombi. Percutaneous catheter ablation was performed without discontinuing anticoagulation. Electroanatomical mapping was performed using the CARTO 3 system and a 20-pole PentaRay catheter (Biosense Webster, Irvine, CA, USA). Voltage mapping was performed during sinus rhythm using a PentaRay catheter before ablation. Fifteen patients were in sinus rhythm before the procedure, and the remaining patients needed external electrical cardioversion to restore sinus rhythm. Circumferential pulmonary vein isolation (PVI) was applied using the technique and method described in previous study [11]. The cavotricuspid isthmus line was ablated if a typical atrial flutter was identified. Electrical cardioversion was performed for patients in which AF remained. Box lesion, also known as posterior wall isolation, was performed at the discretion of the physicians. The bilateral posterior PVI lines, roof, and floor ablation lines encircle the box lesion. Bidirectional conduction block was confirmed by performing differential pacing maneuvers.

Based on previous publications, we defined the presence of fibrosis as a region with a bipolar amplitude 0.5 mV [12, 13]. The extent of atrial fibrosis in this study was classified into normal (no detectable voltage area 0.5 mV), mild fibrosis (proportion of LA surface with severe fibrosis (0.5 mV) of less than 50%); moderate fibrosis (proportion of LA surface with severe fibrosis (0.5 mV) of more than 50%); severe fibrosis (diffuse diseased regions, Supplementary Fig. 1).

The clinical follow-up consisted of physical examinations, transthoracic echocardiography, electrocardiogram, and 24-hour Holter monitoring performed 3 and 6 months after the procedure and every 6 months thereafter. Any cardiac symptoms that presented after ablation were considered an indication for 24-hour Holter monitoring. Recurrence of atrial arrhythmias was defined as any documented electrocardiographic episode of atrial arrhythmia lasting 30 seconds or longer with or without symptoms.

This study had two primary endpoints. The first was recurrent atrial arrhythmias associated with atrial fibrosis. The second primary endpoint was the change in the severity of MR associated with atrial fibrosis.

Continuous variables are presented as the mean $\pm$ standard deviation and were compared using a two-tailed t-test. Categorical variables are presented as frequencies and percentages and were compared using a chi-square or Fisher’s exact test. Continuous variables were compared using a paired t-test for paired samples within groups. A p-value 0.05 was considered to be statistically significant. Statistical analyses were conducted using SPSS software version 26.0 (IBM Corporation, Armonk, NY, USA).

The enrollment procedure for patients is illustrated in Fig. 1. In total, 1045 patients underwent catheter ablation for persistent AF at three centers between 2020 and 2022. After review, 83 patients met our criteria for moderate or severe MR; 75 patients remained for the 1-year clinical follow-up and were included in this study.

Fig. 1.

Flowchart illustrating the mitral regurgitation cohort selections. AF, atrial fibrillation; MR, mitral regurgitation.

Following endocardial 3D voltage mapping, nine patients with mild atrial fibrosis and 32 patients with normal atrial voltage were defined as the mild fibrosis group. Additionally, 12 patients with moderate and 22 with severe atrial fibrosis were defined as the significant fibrosis group. Beyond PVI, 14 patients (18.7%) underwent box lesions.

The baseline clinical and echocardiographic characteristics are presented in Table 1. In contrast to those with mild fibrosis, the group with significant fibrosis had a higher prevalence of congestive HF (New York Heart Association (NYHA) II–III: 76.5% vs. 36.6%, p 0.001), more enlarged atria (LA: 47.2 $\pm$ 5.6 mm vs. 44.1 $\pm$ 4.7 mm, p = 0.012; RA: 46.7 $\pm$ 6.2 mm vs. 41.4 $\pm$ 5.8 mm, p 0.001), and a higher tricuspid regurgitation peak gradient (TRPG, 34.0 $\pm$ 8.2 mmHg vs. 27.2 $\pm$ 7.9 mmHg, p = 0.001). Pacemakers were implanted in two patients due to sick sinus syndrome and in one patient owing to an atrioventricular conduction block.

Within the 1-year follow-up, 28 (37.3%) patients in the entire cohort had atrial arrhythmia recurrences, with 10 having atrial tachycardia (AT). The group with significant fibrosis had a higher incidence of recurrent atrial arrhythmias (55.9% vs. 22.0%, log-rank p = 0.002) than the mild fibrosis group (Fig. 2). In the mild fibrosis group, AF recurred in five patients, three of whom had paroxysmal episodes; three patients had recurrent AT. In the significant fibrosis group, AF recurred in 13 patients, four of whom had paroxysmal AF; the other six had recurrent AT.

Fig. 2.

Cumulative incidence curves of recurrence of atrial arrhythmia.

The atria size decreased in the entire cohort (LA: 41.6 $\pm$ 6.5 mm vs. 45.5 $\pm$ 5.3 mm, p 0.001; RA: 39.9 $\pm$ 9.6 mm vs. 43.8 $\pm$ 6.5 mm, p = 0.008), as well as the severity of MR and TR (Fig. 3).

Fig. 3.

Echocardiographic outcomes in the total cohort. MR (A) and TR (B) severity at baseline and follow-up. MR, mitral regurgitation; TR, tricuspid regurgitation.

At the follow-up visit, the group with significant fibrosis had a more enlarged LA size (44.4 $\pm$ 6.4 mm vs. 39.2 $\pm$ 5.6 mm, p = 0.012) and RA (44.7 $\pm$ 11.2 mm vs. 36.7 $\pm$ 6.5 mm, p 0.001), and more severe TR (Fig. 4B) than the mild fibrosis group. In the significant fibrosis group, the atria size did not decrease after ablation (Fig. 4C,D), and 14 patients (41.2%) had moderate or severe MR at follow-up (Fig. 4A). LA size was not significantly different between the group with recurrent arrhythmias and the sinus rhythm group (42.9 $\pm$ 7.3 mm vs. 40.8 $\pm$ 5.8 mm, p = 0.185). In the multivariable regression analysis, residual moderate and severe MR was associated with female gender [odd ratio (OR) 4.336, 95% CI 1.340 to 14.029] and LA diameter (OR 1.113, 95% CI 1.016 to 1.220).

Fig. 4.

Echocardiographic outcomes in mild and significant fibrosis groups. MR (A) and TR (B) severity at baseline and follow-up. Left atrial (C) and right atrial (D) size at baseline and follow-up between groups. MR, mitral regurgitation; TR, tricuspid regurgitation.

Of the 34 patients with significant fibrosis, one patient with recurrent AF underwent a stroke despite anticoagulant therapy, one patient with recurrent AT had a transient ischemic attack (TIA), and one patient died due to cerebral hemorrhage. Four patients were re-hospitalized for congestive HF; two of them were in sinus rhythm. One patient with mild fibrosis died due to non-cardiovascular disease.

The incidence of significant AFMR was reported in 8.1% of patients with persistent AF, whereas the prevalence for patients with longstanding persistent AF (duration $>$10 years) was 28% [14]. The event-free rate for cardiac death or HF hospitalization was 53% at 24 months, and those with concomitant secondary TR had a worse prognosis, with an event-free rate of 27%. In our group with significant fibrosis, moderate and severe TR was found in 67.6% of patients at baseline and 29.4% at follow-up. The combination of AFMR and TR, which had the poorest prognosis, should receive greater therapeutic attention.

In a study involving hospitalized HF patients with AF, the prevalence of AFMR was 15.9% (30/189), and the all-cause mortality and HF rehospitalization rates over a median period of 273 days were 13.3% and 36.7%, respectively [15]. AFMR patients had a higher risk of cardiac death and HF readmission than HF patients without MR, irrespective of LVEF. Based on these observations, AFMR requires increased attention because of the associated high cardiac event rate and poor prognosis.

The mechanisms underlying AFMR were reported previously. AF induces electrical and structural remodeling of the LA, leading to dilatation, which in turn causes enlargement of the mitral annulus. Mitral annular dilatation, atriogenic leaflet tethering, and insufficient leaflet remodeling also contribute to the pathogenesis of AFMR [1, 16]. Theoretically, restoration of sinus rhythm could reverse atrial remodeling and decrease MR. However, the extent of left atrial fibrosis is associated with a lower success rate following ablation [17].

The potential importance of atrial myopathy in AFMR is exhibited not only in electrical remodeling and the maintenance of AF but also in poor atrial systolic function assessed by strain measurements [4]. In cases of HF with preserved LVEF, atrial dysfunction plays the dominant role in the pathophysiology of functional MR, which reversely increases atrial tension and progressively leads to atrial dysfunction; the bidirectional relationship creates a vicious cycle [4].

The recent 2021 ESC/EACTS guidelines for valvular disease recommend valve surgery or transcatheter edge-to-edge repair (TEER) for patients with severe MR who are consistently symptomatic despite standard medical therapy for HF [18]. Nevertheless, the guidelines did not divide the origin of secondary MR into ventricular and atrial dysfunction. Additionally, the recommendations in the 2020 Japanese guidelines, particularly for AFMR, are similar to the 2021 ESC/EACTS guidelines mentioned above, but with the recommendation that catheter ablation is reasonable for symptomatic patients with persistent AF and severe MR if successful ablation and maintenance of sinus rhythm can be expected [19]. In a retrospective study involving a small number of patients with AF and AFMR, catheter ablation was associated with a lower risk of readmission from HF and stroke compared to those treated with medical therapy [20]. However, the patients with AF recurrence exhibited no reduction in LA size and severity of MR after ablation [3]. Since a considerable proportion of increased remodeled atrial substrate exists in patients with persistent AF, catheter ablation might not be sufficient to reduce atrial size and MR severity through rhythm control, regardless of the ablation strategy. The recently released DACAAF II study revealed that fibrosis-targeted ablation plus PVI, compared with PVI alone, resulted in no significant difference in atrial arrhythmia recurrence [21]. The arrhythmogenic propensity of fibrotic tissue could not be eliminated by ablation, as atrial fibrosis was detected to be progressive and contributed to the recurrence of arrhythmias [22]. Combined with our study, these studies suggest atrial fibrosis may serve as a prognostic surrogate for more advanced disease. Moreover, these studies provide a rationale for investigating the optimal intervention for both atrial reversal remodeling and hemodynamic improvement.

TEER has been proven efficient in several small retrospective studies [23, 24, 25] and a registry study [26], with an acute and long-term reduction in the MR and NYHA functional class. However, current studies cannot definitely determine whether valve intervention is superior to ablation due to the limited sample size and the single-arm or retrospective design. A retrospective study reported that surgical valve repair combined with surgical ablation resulted in less readmission for HF and AF than catheter ablation alone over an 8-year follow-up period [27]. Another retrospective study found that surgical valve repair with concomitant ablation could benefit patients regarding recurrent MR, compared to surgical repair alone [28]. In this setting, well-designed comparative studies involving valve intervention, rhythm control, or combined strategies are needed for treating patients with AFMR and AF.

The retrospective, observational design of our study makes it difficult to establish causal relationships. One of the main limitations was that mitral annular parameters and morphology were not measured by echocardiography or computed tomography (CT) scan, meaning the differences between subgroups could not be examined. Moreover, atrial myopathy involves the LA and the RA, manifested by a dilated RA diameter, TR, atrial tachyarrhythmias, and sinus node dysfunction [29]. Although we demonstrated an enlarged RA and significant TR in the study cohort, there was a lack of RA mapping during the ablation procedure, and we were unable to evaluate the extent of fibrosis in both atria. Further large-scale, prospective cohort studies are required to confirm and extend our findings on the relationships between AF, atrial fibrosis, and MR.