Effects of rhythm control on left atrial structure remodeling in atrial fibrillation and heart failure with preserved ejection fraction

Background The benefits of rhythm control for atrial fibrillation (AF) in heart failure with preserved ejection fraction (HFpEF) have not been conclusively determined. We assessed the effects of rhythm control on left atrial (LA) structure remodeling and prognosis in patients with AF and HFpEF. Methods This was a retrospective, real-world, observational study involving patients diagnosed with AF and HFpEF. The cohort was divided into rhythm-control and rate-control groups depending on their treatment strategies. The primary outcomes were all-cause mortality, rehospitalization for any cause, HF-related rehospitalization, and stroke. Differences in follow-up LA structure parameters were also analyzed. Results Compared to the rate-control group, patients in the rhythm-control group had a lower risk of HF-related rehospitalization even after adjusting for potential confounders (adjusted HR 0.605, 95% CI 0.413–0.887, p = 0.010). Moreover, rhythm-control therapy led to marked reductions in LA echocardiographic indicators and a higher proportion of LA reverse remodeling (LARR). Conclusions Rhythm-control therapy reverses LA structure remodeling and is associated with improved clinical outcomes; therefore, it is an optimal treatment approach for AF in HFpEF patients.


Introduction
Globally, approximately 64.3 million people are suffering from heart failure (HF), among whom, heart failure with preserved ejection fraction (HFpEF) accounts for about 36% of the total HF spectrum (1).Atrial fibrillation (AF) is one of the most common complications in HFpEF patients.The Olmsted County cohort study revealed that approximately two-thirds of HFpEF cases had AF during the disease process (2).In another study, AF patients were 4.8 times more likely to develop HFpEF than those without AF (3).These two diseases often share common risk factors, including advanced age, female sex, anemia, and comorbidities such as hypertension, obesity, chronic renal insufficiency, atherosclerosis, and sleep apnea.Additionally, AF and HFpEF interact and promote their progress through cardiac structure and electrophysiology remodeling, neurohumoral activation, inflammation, and other pathogenic mechanisms, resulting in a vicious cycle (4).
The latest therapeutic guidelines recommend that rhythm control therapy, particularly catheter ablation (CA), is a potentially effective strategy for treating symptomatic AF and HF (5,6).There is sufficient evidence that patients with heart failure with reduced ejection failure (HFrEF) patients and AF who were subjected to catheter ablation (CA) to restore sinus rhythm (SR) exhibited better prognostic outcomes compared to those treated with rate-control therapy (7)(8)(9)(10)(11)(12).However, little attention has been paid to HFpEF, and a limited number of studies have been conducted to assess the efficacy of rhythm control therapy (CA, electrocardioversion, or surgical ablation) for HFpEF and AF.According to the available evidence, HFpEF populations may benefit from rhythm control therapies.Thus, to verify it, we investigated the effect of rhythm-control therapy on prognosis and left atrial remodeling in patients with HFpEF and AF.

Study population Definitions
The diagnosis of HFpEF and AF is determined in accordance with the guideline of the European Society of Cardiology (6).The diagnostic criteria for HFpEF included: (i) typical signs and symptoms of HF; (ii) LVEF ≥50%; and (iii) evidence of spontaneous or provokable increased LV filling pressures (elevated brain natriuretic peptide, non-invasive and invasive hemodynamic measurements).Clinically, AF is defined as ≥30 s recorded in a standard 12-lead ECG or single-lead ECG, with no identifiable repetitive P waves and irregular RR intervals when atrioventricular conduction was not impaired.

Clinical data
Demographics, the New York Heart Association (NYHA) functional class, co-morbidities, laboratory tests, and information on drug therapies were collected.Some valuable echocardiographic parameters listed as follows were also recorded.Mitral Doppler early velocity/mitral annular early velocity (E/e'), the ratio of peak velocity of mitral inflow during early diastole (E), and over the average septal and lateral mitral annular early diastolic peak velocities (e'), reflects left ventricular (LV) stiffness and fibrosis.Left atrial volume index (LAVI), an essential indicator of chronic LA remodeling, is measured at end-systole and indexed to body surface area.Left ventricular mass index (LVMI) and relative wall thickness (RWT) are important markers of global LV remodeling or hypertrophy.LVMI is calculated using the Cube formula and indexed to body surface area.RWT is twice the LV posterior wall thickness divided by the LV internal diameter at end-diastole.Tricuspid regurgitation (TR) peak velocity is considered to be an indirect marker of LV diastolic dysfunctions (13).Left atrial reverse remodeling (LARR), a vital index of LA structure reverse remodeling, is defined by a ≥15% reduction in LA end-systolic volume (LAESV) (14).

Endpoints on follow-up
The primary clinical outcomes in this study were all-cause mortality, rehospitalization for any cause, HF-related readmission, and stroke.Differences in follow-up LA structure parameters were calculated and compared.The time interval between the two echocardiogram examinations was at least 3 months.Postoperatively, all participants received at least three outpatient visits after ablation (3,6, and 12 months within 1 year after surgery and once every 6 months after 1 year).During every patient visit or when they felt uncomfortable, an ECG or Holter examination was performed to monitor atrial arrhythmia.A telephone follow-up was made once they did not attend their scheduled appointments.The deadline for follow-up was December 30, 2021.

Statistical analysis
Normally or non-normally distributed continuous variables were expressed as mean ± standard deviation or median (interquartile range).They were analyzed using an independent-sample t-test or Mann-Whitney's U-test.Categorical variables were presented as percentages (%), and differences were assessed using a chi-square test.Cumulative incidences of prespecified endpoints were calculated through the Kaplan-Meier curve, with log-rank tests comparing the differences.Cox proportional hazard models were used to compare the risks of adverse events after adjustments, with hazard ratios (HR) and 95% confidence intervals (95% CI) presented.All factors with p < 0.05 in the univariate Cox analysis were entered into the multivariate model.p < 0.05 was the threshold for statistical significance.The Statistical Package for Social Sciences (SPSS) version 26.0 was used for analysis.

Baseline characteristics
A total of 596 patients with HFpEF and AF were initially enrolled.Of these, 225 cases were assigned to the rhythm-control group and 371 to the rate-control group.A 1:1 optimal PS matching with a 0.02 caliper and no replacement was conducted to reduce selection bias and potential confounding factors, yielding 177 successfully matched pairs from the entire cohort.

LA structure parameters on follow-up
Differences in follow-up echocardiographic parameters were compared between the two groups (Table 2).Compared with the rate-control group, the rhythm-control group showed more significant reductions in LA diameter (LAD) [rhythm control: from 45.88 ± 0.73 to 44.94 ± 0.68 vs.rate control: from 46.65 ± 0.69 to 48.41 ± 0.79, p = 0.001], LA superior-inferior diameter The flow chart of the study population.2).Moreover, rhythm-control therapy was also associated with a higher incidence of LA reverse remodeling than traditional rate-control therapy [44 (24.90%) vs. 13 (7.30%),p < 0.001] (Figure 3).Other indicators, such as IVS, LVPWT, LVEDD, LVMI, E/e', TRV, and RWT, were similar between the two groups at follow-up.

Adverse outcomes on follow-up
During follow-up, 25 patients died [rate-control: n = 15 (8.5%) vs. rhythm-control: n = 10 (5.6%)], and 42 cases developed stroke [rate-control: n = 27 (15.3%)vs. rhythmcontrol: n = 15 (8.5%)].A total of 223 patients were rehospitalized [rate-control: n = 111 (62.7%) vs. rhythm-control: n = 112 (63.3%)], of whom 154 were readmitted for worsening HF [rate-control: n = 88 (45.8%) vs. rhythm-control: n = 66 (37.3%)].Kaplan-Meier analysis showed that patients in the rhythm-control group had lower incidences of HF-related rehospitalization ( p = 0.003) and stroke ( p = 0.027) when compared to those in the rate-control group (Figure 4).Cox regression analysis also demonstrated that patients treated with rhythm-control therapy experienced lower risks of HF-related rehospitalization after adjusting for three potential confounders (adjusted HR 0.605, 95% CI 0.413-0.887,p = 0.010) but not with stroke (adjusted HR 0.589, 95% CI 0.279-1.242,p = 0.164) (Table 3).Differences in the risks of hospitalization for any cause and all-cause mortality among the two groups were insignificant either before or after adjustment (Table 3).Survival analysis was also performed to compare the efficacy of different rhythm control strategies for AF and HFpEF, aiming to identify a convenient and inexpensive approach for individuals unsuitable for CA.Surprisingly, CA did not show a remarkable superiority in reducing the risks of preset endpoints over ECV (data not shown).Changes in left atrial structure parameters between baseline and follow-up in rhythm and rate-control groups after PS-matching.

Discussion
Among HFpEF and AF patients, rhythm-control treatment effectively reduced the risk of adverse endpoint.It significantly contributed to improved LA size, characterized by decreased LA volume and a higher rate of LA reverse remodeling compared to rate-control treatment.Some extensive randomized controlled studies, such as CASTLE-AF, CABANA-HF, and EAST-AFNET4 trials, established that CA therapy for AF and HF, especially HFrEF, was superior to conventional ventricular rate medications in enhancing cardiac functions, improving the quality of life, and reducing the risks of death and admission (7,8,10,12).Based on this, the latest guideline recommends that patients with Left atrial reverse remodeling in the rhythm and rate-control groups after PS-matching.Risks of adverse events between rhythm and rate-control groups after PS-matching.Event-free proportions from (A) all-cause mortality, (B) HF-related rehospitalization, (C) rehospitalization for any cause, (D) stroke.A sizeable retrospective study revealed that compared with the rate control therapy, rhythm control was more effective in decreasing all-cause mortality in elderly AF and HFpEF patients after 1-year follow-up (25).Another study involving 85 HFpEF and AF patients showed that CA markedly reduced the rehospitalization risk for HF (26).Two metaanalyses showed that CA reduced the rehospitalization rate for HF and all-cause mortality in HFpEF and AF patients (27,28).Our findings were consistent with these studies, further elucidating the clinical benefits of rhythm control in decreasing the incidence of HF-related rehospitalization for HFpEF and AF.Two other studies reported that CA decreased the risk of ischemic stroke [HR 0.69, 95% CI (0.51-0.93) and RR 0.22; CI (0.09-0.56)] (29, 30).Our study also confirmed that rhythm treatment was associated with a lower stroke hazard, even if there was no statistical difference after adjustment.However, no significant reduction in the risks of all-cause mortality and rehospitalization was observed for rhythm treatment in the study.This may be due to the small sample size, lower mortality rate, and relatively shorter follow-up period.Thus, large clinical trials are necessary to be carried out to detect the impact of rhythm treatments in these patients.HFpEF and AF are closely associated with atrial failure, defined as any atrial dysfunction resulting in impaired HF-related symptoms and worsening life expectancy in the absence of significant valvular or ventricular abnormalities, which was proposed in 2020 (31).LA structure remodeling, manifested by LA dilation, is one of the most critical underlying mechanisms of atrial failure, leading to increased incidences of HFpEF and AF (4,32).In turn, AF induces oxidative stress, atrial enlargement, calcium overload, inflammation, immune cell infiltration, microRNA expression, and myofibroblast activation, resulting in atrial remodeling, increasing the risk of HF and, ultimately, a vicious cycle (33)(34)(35)(36).Thus, timely and effective intervention of AF may promote LA reverse remodeling and improve clinical endpoints.In this study, we analyzed some LA indicators and found that they were significantly decreased after rhythm-control therapy, with a higher proportion of LARR.These findings showed that rhythm treatment had a more favorable effect on reversing LA structure remodeling.Rhythm treatment may improve LA structure remodeling, greatly stabilizing hemodynamics and preventing HF deterioration.
In addition to CA, left atrial appendage closure (LAAC) is an effective strategy to reduce the risk of ischemic stroke/TIA/ systemic embolism during follow-up (37).The combination of CA and LAAC has been proven beneficial to the long-term prognosis of patients with AF.A meta-analysis demonstrated that ablation with the vein of Marshall ethanol infusion (VOM-ABL) was associated with higher rates of successful mitral isthmus (MI) block and long-term freedom from AF/AT and was comparable safety compared with ablation alone in patients with AF followed over 1 year or more (38).Moreover, a significant proportion of patients cannot receive CA treatment due to various reasons (old age, over-dilated LA, multiple comorbidities, and worse cardiac functions) in clinical practice.Therefore, it is necessary to identify an alternative treatment strategy for this population.Interestingly, patients who received ECV and those subjected to CA exhibited comparable benefits in reducing the incidence of adverse events, indicating that ECV may be an effective alternative for those who cannot tolerate CA or may benefit less.These recent advancements also needed to be verified in the HFpEF population through large multi-center trials.
We admitted that there were several limitations involved in this study.Firstly, the study was neither a randomized controlled nor a large multi-center.Thus, our findings may not apply to all populations.Secondly, all participants were retrospectively identified; selection bias was inevitable.Thirdly, as there was no significant difference in the number of cases between the two groups, conducting a successful 1:1 match based on all potential confounders is unavailable, resulting in some difference in baseline data among the two groups.Fourthly, owing to financial factors and patients' wishes, implanting the cardiac monitor to record EKG for all patients may be difficult.Thus, some recurrent atrial arrhythmia might be missed.Lastly, the endpoints were mainly obtained by telephone interview, which may cause bias.
In conclusion, rhythm-control therapy reverses LA structure remodeling and is associated with improved clinical outcomes; therefore, it is an optimal therapeutic option for AF in HFpEF patients.

TABLE 1
Baseline characteristics between rhythm and rate-control groups before and after PS-matching.

TABLE 2
Echocardiographic parameters between rhythm and rate-control groups after PS-matching on follow-up.
(23)(20)(21)(22)e pieces of evidence have shown that HFpEF significantly differs from HFrEF in pathogenesis, clinical manifestations, and prognosis(16)(17)(18).Patients with HFpEF are more female, have lower NP levels, and show a higher prevalence of atrial fibrillation and hypertension but a lower incidence of ischemic heart disease(16).Several large RCT trials have reported that, unlike HFrEF, most HFpEF patients respond poorly to conventional HF medical therapy (ACEI/ARB/ARNI, β-blocker, MRA) apart from loop diuretics and sodium-dependent glucose transporters-2 inhibitors(19)(20)(21)(22).Recent basic research found that pyroptosis of adipocytes from epicardial adipose tissue was deeply involved in the myocardial damage in a mouse model of HFpEF,(23)implying the great importance of non-cardiac mechanisms in HFpEF.Therefore, there should also be more evidence of the clinical benefits of CA in HFpEF.Black-Maier et al. reported that CA significantly improved NYHA class and alleviated the symptoms in HFpEF and AF patients (24).

TABLE 3
Risks of adverse events between rhythm and rate-control groups after PS-matching.