Efficacy and safety of catheter ablation for atrial fibrillation in patients with heart failure with preserved ejection fraction: a systematic review and meta-analysis

Background Catheter ablation (CA) effectively treats atrial fibrillation (AF) in heart failure (HF) with reduced ejection fraction (HFrEF), improving clinical outcomes. However, its benefits for AF patients with heart failure with preserved ejection fraction (HFpEF) are still unclear. Methods We systematically searched PubMed, Embase, Web of Science, the Cochrane Library, and Scopus for studies investigating outcomes of CA in AF patients with HFpEF. Efficacy indicators included freedom from AF and antiarrhythmic drugs (AAD) free AF elimination. Safety indicators comprised total complications, HF admission, all-cause admission, and all-cause mortality. Sixteen studies with 20,796 patients included in our research. Results The comprehensive analysis demonstrated that, when comparing CA with medical therapy in HFpEF, no significant differences were observed in terms of HF admissions, all-cause admissions, and all-cause mortality [(OR: 0.42; 95% CI: 0.12–1.51, P = 0.19), (HR: 0.78; 95% CI: 0.48–1.27, P = 0.31), and (OR: 1.10; 95% CI: 0.83–1.44, P = 0.51)], while freedom from AF was significantly higher in CA (OR: 5.88; 95% CI: 2.99–11.54, P < 0.00001). Compared with HFrEF, CA in HFpEF showed similar rates of freedom from AF, AAD-free AF elimination, total complications, and all-cause admission were similar [(OR:0.91; 95% CI: 0.71,1.17, P =0.47), (OR: 0.97; 95% CI: 0.50–1.86, P = 0.93), (OR: 1.27; 95% CI: 0.47–3.41, P = 0.64), (OR: 1.11; 95% CI: 0.72, 1.73; P = 0.63)]. However, CA in HFpEF was associated with lower rates of HF admission and all-cause mortality [(OR: 0.35; 95% CI: 0.20, 0.60; P = 0.0002), (OR: 0.40; 95% CI: 0.18, 0.85; P = 0.02)]. Compared with patients without HF, CA in HFpEF patients exhibited lower rates of AAD-free AF elimination (OR: 0.48; 95% CI: 0.30, 0.75; P = 0.001). However, their rates of freedom from AF and total complications were similar [(OR: 0.70; 95% CI: 0.48, 1.02; P = 0.06), (OR: 0.60; 95% CI: 0.19, 1.90; P = 0.38)]. Conclusion This meta-analysis conducted provided a comprehensive evaluation of the efficacy and safety of CA in patients with AF and HFpEF. The results suggest that CA may represent a valuable treatment strategy for patients with AF and HFpEF. Systematic Review Registration https://www.crd.york.ac.uk/PROSPERO/#recordDetails, identifier (CRD42024514169).


Introduction
Constituting at least 50% of prevalent instances, HFpEF represents a substantial share of global heart failure cases and is associated with an unfavorable prognosis (1,2).Effectively addressing HFpEF requires a holistic strategy that prioritizes the screening and management of underlying causes and associated comorbidities in patients (3,4).Due to the shared risk factors and common pathophysiological processes between AF and HF, AF frequently coexists in HFpEF patients.The prevalence of AF in HFpEF patients ranges from 40%-60%.Moreover, it is associated with heightened morbidity, thromboembolic susceptibility, hospitalization frequencies, and mortality rates (5)(6)(7)(8)(9).Recently, there has been growing interest in determining the optimal therapeutic approach for patients with AF and HFpEF.
CA is an effective and relatively safe therapeutic option for AF (10).It has increasingly become a successful and common strategy for managing symptomatic AF (11).HF may heighten the likelihood of complications related to CA.While researches have indicated that CA interventions can lead to a reduction in AF burden and confer benefits in terms of decreased rehospitalization and mortality rates among AF patients with HFrEF.However, there is scarce data on whether CA treatment for AF can offer clinical advantages for HFpEF patients (12,13).
To assess the efficacy, a single-arm meta-analysis ( 14) aggregated multiple studies on CA utilization in patients with AF and HFpEF.The analysis indicated that CA can be beneficial in maintaining sinus rate.However, being a single-arm study, it is subject to certain limitations, including the absence of a control group and notable heterogeneity among the included studies.Up to date, there were few published studies and meta-analyses that related to the outcomes of CA in patients with AF and HFpEF, the results were inconsistent, and the available studies were inadequate to ascertain the superiority of catheter ablation in these patient cohorts.This meta-analysis aims to evaluate both the efficacy and safety of CA compared to medical therapy for patients with AF and HFpEF.Additionally, the study sought to assess the efficacy and safety of CA in patients with AF, encompassing both HFpEF and HFrEF, as well as individuals with HFpEF and without HF.

Literature search
In accordance with the PRISMA 2020 statement (15), our evidence-based analysis was conducted and prospectively registered in PROSPERO (CRD42024514169).The PRISMA 2020 checklist guided our methodology, ensuring transparency and rigor throughout the review process.Up to December 18, 2023, our systematic literature search comprehensively covered five databases, including PubMed, Embase, Scopus, Cochrane Library, and Web of Science.
The principal objective of this study was to systematically identify and review research studies examining the efficacy and safety of CA in the context of AF among patients with HFpEF.Only English-language publications were considered.The following terms were used to search the databases: "Atrial fibrillation", "AF", "Heart Failure", "HF", "heart failure with preserved ejection fraction", "HFpEF", "Catheter ablation" and "CA."A detailed outline of the search strategy is provided in Supplementary Table S1.
Additionally, all the eligible articles were manually reviewed to ensure comprehensiveness.Two investigators conducted the search and assessment of included literature independently.Any discrepancies encountered during the literature search or assessment process were resolved through mutual consensus and discussion.

Identification of eligible studies
The criteria for eligible studies included: (1) randomized controlled, cohort, or case-control study design; (2) studies involving AF patients with HFpEF, HFrEF, and without HF; (3) studies examining CA and medical therapy(rhythm control using AAD) for AF patients with HFpEF, CA for AF patients with HFpEF and HFrEF (EF 40%-49% included in the HFrEF group), and CA for AF patients with HFpEF and without HF; (4) Studies that provided dependable information on outcomes (freedom from AF,AAD-free AF elimination, total complications, HF admission, all-cause admission, and all-cause mortality); (5) Studies had to be published in a peer-reviewed scientific journal.
We excluded reviews, letters, editorial comments, case reports, conference abstracts, unpublished articles, and articles not written in English.

Endpoints
The primary endpoint was efficacy-freedom from AF and AAD-free AF elimination.Freedom from AF was defined as absence of any symptomatic or asymptomatic atrial arrhythmia lasting more than 30 s after completing the blanking period (3 months) after CA or medical treatment.AAD-free AF elimination refered to achieving the elimination of AF without the use of AAD by the end of follow-up.
The secondary endpoint was safety outcomes.Safety outcomes included total complications after CA, HF admission, all-cause admission, and all-cause mortality.Total complications varied among studies, encompassing cardiac perforation/tamponade, access site/vascular complications, pericarditis, esophageal atrial fistula, pulmonary vein stenosis, phrenic nerve injury, acute heart failure, stroke/transient ischemic attack, air embolism, and prolonged hospitalization.
investigator, Shenghong Feng, to achieve a conclusive resolution.The data extracted from the included studies encompassed the first author's name, publication year, country, study period, study design, sample size, age, body mass index (BMI), comorbidities, AAD usage, AF type, duration of AF before intervention, follow-up duration, left atrium (LA) volume, E/E', left ventricular end-diastolic dimension (LVEDd), procedure time, the primary endpoint and the secondary endpoint.When continuous variables were presented as median with range or interquartile range in the study, we derived the mean ± standard deviation using a validated mathematical approach (16,17).In instances where data were either missing or unreported, we reached out to the corresponding authors to acquire complete data sets, if feasible.

Quality assessment
Two reviewers, Xiaomei Chen and Xuge Zhang, independently conducted the quality assessment of included studies.Randomized controlled trials underwent evaluation using the Cochrane risk of bias tool (18).Meanwhile, observational studies were assessed using the Newcastle-Ottawa Scale (NOS) (19), which employs 3 domains of selection, comparability and outcome/exposure.

Statistical analysis
The evidence synthesis utilized Review Manager version 5.4 (Cochrane Collaboration, Oxford, UK).Continuous variables were compared using weighted mean differences (WMD), while dichotomous variables were analyzed with odds ratios (OR) or hazard ratio (HR), both accompanied by 95% confidence intervals (CIs).Heterogeneity was assessed using the chi-squared (χ 2 ) test (Cochran's Q) and inconsistency index (I 2 ), with significant heterogeneity defined as χ 2 P-value < 0.05 or I 2 > 50%.Considering the heterogeneity among different studies, all data synthesis was performed using a random-effects model.Subgroup or sensitivity analyses were performed to evaluate the influence of individual studies on outcomes exhibiting significant heterogeneity.To examine publication bias, funnel plots were visually assessed using Review Manager version 5.4, while Egger's regression tests were employed with Stata version 17.0 (StataMP-64) for outcomes involving three or more studies.A significance threshold of P < 0.05 denoted the presence of statistically significant publication bias.

Literature search and study characteristics
The systematic literature search identified 4,389 related studies, with contributions from various databases: 210 from PubMed, 2,028 from Embase, 1,562 from Scopus, 353 from the Cochrane Library, and 235 from Web of Science.Following initial screening based on title and abstract, 3,227 articles underwent further evaluation.Subsequently, the complete texts of 43 articles were carefully reviewed.Among these, 21 conference abstracts and 2 articles (20, 21) with overlapping data were excluded, along with 2 studies (22, 23) lacking available data for analysis.Ultimately, sixteen articles, comprising a total of 20,796 patients, were included in the meta-analysis.These comprised 2 prospective cohort studies (24, 25), 13 retrospective cohort studies (26)(27)(28)(29)(30)(31)(32)(33)(34)(35)(36)(37)(38), and 1 prospective randomized study (39).Detailed study and patient characteristics are provided in Table 1.Among the included patients, 2,935 (14.1%) of 20,796 AF patients with HFpEF underwent CA, while 15,903 (76.5%) received medical therapy.Additionally, 794 (3.8%) of 20,796 AF patients with HFrEF and 1,164 (5.6%) of 20,796 AF patients without HF underwent CA.The study's flowchart is illustrated in Figure 1, while a comprehensive risk of bias analysis for the included studies is available in Supplementary Table S2 and Figure S1.

Sensitivity analysis
A one-way sensitivity analysis was performed to evaluate the influence of individual studies on the aggregated results, with the outcome consistently demonstrating coherence.Regardless of the exclusion of any individual study concerning HF admission, the analysis revealed that the recalculated combined odds ratio remained consistently stable (Figure 3).
3.3 CA for patients with HFpEF vs. HFrEF

Freedom from AF
Synthesized from a compilation of 10 studies, the data on freedom from AF involve a total of 2,523 patients, with 1,729 having HFpEF and 794 with HFrEF.No significant difference in freedom from AF were observed (OR: 0.91; 95% CI: 0.71, 1.17, P = 0.47), with no statistically heterogeneity observed (I 2 = 39%; Figure 4A).While a slight visual indication of publication bias was noted in Figure 5A, Egger's test did not reveal any statistically significant evidence of publication bias (P = 0.650).

All-cause mortality
Analysis of all-cause mortality was conducted in 6 studies with 1,786 patients (1,267 HFpEF vs. 519 HFrEF).Pooled analysis indicated that CA in HFpEF resulted in a significant reduction in all-cause mortality (OR: 0.40; 95% CI: 0.18, 0.85; P = 0.02), with no statistically heterogeneity (I 2 = 39%; Figure 4F).While a slight indication of publication bias was observed in Figure 5F, Egger's test did not identify publication bias (P = 0.364).

Sensitivity analysis
One-way sensitivity analyses were conducted to evaluate the influence of individual studies on different outcomes, including AF recurrence, AAD-free AF elimination, total complications, HF admission, all-cause admission, and all-cause mortality.This evaluation aimed to determine the consistency of results by systematically removing each study.Sensitivity analyses indicated that the findings remained consistent across all outcomes, as illustrated in Figures 6A-F.

Sensitivity analysis
During sensitivity analysis, the results remained consistent even after excluding each individual study (Figure 9).Presently, research on HF and AF is increasingly focusing on HFpEF and AF.Not only is the prevalence of HFpEF gradually increasing, surpassing that of HFrEF, but also the incidence of AF in HFpEF is higher than in HFrEF.This coexistence of HFpEF and AF poses a greater risk.CA is acknowledged as an effective treatment for AF and AF with HFrEF.However, the potential benefits of CA for patients with AF and HFpEF remain uncertain.
To evaluate the outcomes of CA in patients with AF and HFpEF, our meta-analysis systematically synthesized findings from sixteen published studies.The key findings are outlined below: (a) CA significantly improves freedom from AF in HFpEF patients, while demonstrating similar rates of HF admission, allcause admission, and all-cause mortality compared to medical therapy.(b) Patients with HFpEF derive comparable benefits from CA as HFrEF patients, with lower rates of HF admission and all-cause mortality.Additionally, HFpEF patients experience similar rates of freedom from AF, AAD-free AF elimination,  total complications, and all-cause admission compared to HFrEF patients.(c) Although the AAD-free AF elimination was lower after CA in HFpEF than in those without HF, patients with HFpEF benefited from CA similarly to those without HF, with a similar freedom from AF and total complications.In achieving rhythm control among AF patients with HFpEF, our study, consistent with the findings of Gu et al. (40), demonstrates that CA surpasses medical therapy.Aligned with Aldaas et al.'s study (41), our systematic meta-analysis underscores that, in patients with HFpEF, CA yields results in terms of freedom from AF comparable to those observed in patients with HFrEF.However, compared to patients without HF, those with HFpEF exhibit lower freedom from AF.It is noteworthy that risk factors for HFpEF, such as increased left atrial diameter, substrate remodeling, arterial hypertension, and female sex, are also associated with lower freedom from AF rates of AF after CA (42).
Previous studies have demonstrated the safety of performing CA in patients with AF, even in the context of HFpEF (40,43).In our study, the total complications are comparable among HFpEF and HFrEF, as well as between HFpEF and those without HF.Within the HFpEF and HFrEF groups, a higher incidence of overall complications was observed in prospective studies, according to the subgroup analysis based on study design.This observation may be attributed to the inclusion of only 2 studies in the prospective design, underscoring the need for further exploration through additional studies.
Reaffirming previous findings (41), our study contributes additional evidence supporting the association between CA involvement in AF patients with HFpEF and a notably diminished risk of all-cause mortality when compared to individuals with HFrEF.Furthermore, our findings also reveal that CA in AF patients with HFpEF results in a lower rate of HF admission compared to those with HFrEF.However, the reason for this difference is not yet clear.Since our study was a post hoc analysis, further insights will be provided by large-scale prospective studies.
CA for AF involves various energy modalities, primarily including radiofrequency ablation and cryoballoon ablation (10).Previous studies have demonstrated that both radiofrequency ablation and cryoballoon ablation are safe and effective for AF patients.However, some studies have indicated that patients undergoing radiofrequency ablation may experience a higher incidence of longterm pulmonary hypertension (44).The majority of studies included in this analysis did not specify the type of energy used for CA.Therefore, more detailed and well-designed studies are needed to elucidate the long-term complications associated with different energy modalities in patients with AF and HFpEF.
The correlation between AF and HF arises from complex interplay of pathophysiological mechanisms.The risk factors for AF and HF exhibit significant overlap, for example, age, obesity, smoking, alcohol consumption, hypertension, diabetes, chronic kidney disease, obstructive sleep apnea, and coronary artery disease (26,(45)(46)(47).Moreover, they have a mutual tendency to precipitate each other, and over time, these conditions may perpetuate or worsen one another in complementary ways.
These identified risk factors contribute to the alteration of both atrial and ventricular function, primarily through inducing inflammatory processes, fibrotic changes, hemodynamic stress, and ischemic events.Consequently, they lead to extensive structural, mechanical, and electrophysiological remodeling (48,49).In the context of AF, it is important to recognize its dual role in the development of HF.Initially, AF can play a direct role in advancing or intensifying HF by compromising atrial contraction and disrupting the coordination between atrial and ventricular contractions.This often leads to diminished cardiac  Funnel plots of freedom from AF. Sensitivity analysis of freedom from AF.Additionally, histopathological studies have elucidated that interstitial fibrosis, increased intercellular gaps, myofibrillar loss, and reduced nuclear density contribute to atrial structural remodeling, which is associated with the persistence of atrial fibrillation and its recurrence post-ablation (51,52).Observational studies suggest that restoring sinus rhythm via CA may modulate the disease progression of AF and HFpEF by interrupting the interdependent perpetuation of AF and HFpEF.CA procedures have shown significant results, such as decreased natriuretic peptide levels, reduced left atrial dimensions, lowered pulmonary capillary wedge pressure during exercise and at rest, increased peak cardiac output, improved New York Heart Association functional capacity, and potential partial resolution of HFpEF (36,37,53,54).In line with a causal relationship, the resolution of HFpEF exhibited a strong correlation with the absence of arrhythmia recurrence following ablation.
Our meta-analysis exhibited several limitations.Firstly, only one randomized controlled trial was included in the analysis.Secondly, the number of studies within each category of analysis is relatively small.In particular, in the category comparing CA to medical therapy for patients with HFpEF, only two or three studies were included.Thirdly, the studies examined did not provide separate data for paroxysmal atrial fibrillation and persistent atrial fibrillation, thus preventing subgroup analyses for these conditions.Fourthly, most of the studies examined did not specify the type of energy used in CA, i.e., thermal (radiofrequency ablation and cryoballoon ablation) or non-thermal (pulsed field ablation), thus precluding further analysis of complications associated with different energy types.Fifthly, the median follow-up duration of the included studies was 23.6 months, highlighting the need for further research to assess the longer-term efficacy of CA.Lastly, publication bias could not be entirely ruled out, as only published data were included, contributing to potential bias.Therefore, it is imperative to conduct more well-designed and large-scale randomized controlled trials to validate these findings.

Conclusions
This meta-analysis comprehensively evaluated the efficacy and safety of CA in patients with AF and HFpEF.The results of the analysis revealed that CA demonstrates comparable safety and effectiveness to medical therapy.Moreover, when compared to patients with HFrEF, CA for HFpEF not only shows similar effectiveness but also leads to reduced rates of HF hospitalizations and all-cause mortality.Despite a lower efficacy of CA in treating HFpEF patients compared to those without HF, its safety profile remains comparable.These findings suggest that CA could serve as an effective and safe strategy for managing patients with AF and HFpEF.

FIGURE 1 Flowchart
FIGURE 1Flowchart of the systematic search and selection process.

FIGURE 3
FIGURE 3Sensitivity analysis of HF admission.

FIGURE 4 Forest
FIGURE 4Forest plots of outcomes: (A) freedom from AF (B) AAD-free AF elimination (C) total complications (D) HF admission (E) all-cause admission (F) allcause mortality.

FIGURE 6
FIGURE 6Sensitivity analysis of outcomes: (A) freedom from AF (B) AAD-free AF elimination (C) total complications (D) HF admission (E) all-cause admission (F) all-cause mortality.

FIGURE 7 Forest
FIGURE 7Forest plots of outcomes: (A) freedom from AF (B) AAD-free AF elimination (C) total complications.

TABLE 1
Baseline characteristics of include studies and methodological assessment.

Table 2
Subgroup analysis.(A).Subgroup analysis of CA for patients with HFpEF vs. HFrEF.
AF, atrial fibrillation; CA, catheter ablation; HF, heart failure; HFpEF, heart failure with preserved ejection fraction; HFrEF, heart failure with reduced ejection fraction; OR, odds ratio; CI, confidence interval; AAD-free AF elimination, antiarrhythmic drugs free atrial fibrillation elimination.(B)Subgroup analysis of CA for patients with HFpEF vs. without HF.