Non-alcoholic fatty liver disease as a predictor of atrial fibrillation recurrence following a retrospective study

Authors: Zhe Wang, Fangyuan Luo, Yuekun Zhang, Yafei Zhai, Xiaojie Chen, Liping Sun, Yingwei Chen

Abstract

Background

Epidemiological evidence suggests an association between non-alcoholic fatty liver disease (NAFLD) and incident atrial fibrillation (AF); however, the magnitude of this association and its prognostic value in predicting the recurrence of AF after radiofrequency catheter ablation (RFCA) have not been fully characterized. The present study was designed to elucidate the complex interplay between NAFLD and the risk of AF recurrence after ablation.

Methods

A total of 1,182 patients with AF who underwent initial RFCA from June 2018 to December 2022 at the First Affiliated Hospital of Zhengzhou University were included in this retrospective cohort study. The Kaplan-Meier method was used to plot AF recurrence curves after ablation. Multivariable Cox models were then used to examine the associations between NAFLD and the recurrence of AF. Analyses were also conducted to assess whether the predictive effect of NAFLD was consistent across different subgroups.

Results

Over a 1-year follow-up period, 30.1% of the patients experienced recurrent AF. The multivariable Cox analysis revealed that NAFLD was an independent risk factor for the recurrence of AF after controlling for model 2 (hazard ratio =1.37, 95% confidence 1.10–1.70, P=0.005). These correlations remained statistically significant across various models. Further, incorporating NAFLD in the fully adjusted basic risk model significantly increased the ability of the model to predict AF recurrence, with the C-statistic increasing from 0.672 to 0.686 (P=0.03). Additionally, diabetes mellitus (DM) (P value for interaction =0.049) and female sex (P value for interaction =0.02) had a statistically significant interactive effect with NAFLD in predicting the recurrence of AF.

Conclusions

NAFLD was found to be independently associated with the recurrence of AF after ablation. Moreover, the AF recurrence rate after RFCA was higher in the NAFLD patients who had DM or were female. The study showed that NAFLD may serve as a dependable marker for assessing AF recurrence risk in clinical practice.

Introduction

Non-alcoholic fatty liver disease (NAFLD) has become one of the most common chronic liver diseases. Epidemiological studies estimate that NAFLD affects nearly 25% of the adult population worldwide (1). NAFLD is increasingly recognized as a multisystem disease (2). NAFLD progression has been found to be associated with metabolic disorders such as impaired glucose metabolism, elevated blood pressure, and reduced insulin sensitivity, the latter of which contributes significantly to increased cardiac electrical instability (3,4). NAFLD has been shown to be associated with a dose-dependent increase in the incidence of new-onset atrial fibrillation (AF), with a particularly heightened risk observed in young adults (aged 20–29 years) (5).

Recent studies have identified several key mechanisms linking NAFLD to AF, including autonomic dysfunction, systemic inflammation, and impaired cardiac relaxation (6,7). However, further research is required to better understand the intricate relationship between NAFLD and the recurrence of AF after ablation, and to determine whether NAFLD is a critical prognostic factor for AF. Given the relationship between metabolic abnormalities and NAFLD, this study aimed to explore the effect of NAFLD on the recurrence of AF after radiofrequency catheter ablation (RFCA). We present this article in accordance with the STROBE reporting checklist (available at https://cdt.amegroups.com/article/view/10.21037/cdt-2025-207/rc).

Methods

Study design and populations

This retrospective observational study was conducted from June 2018 to December 2022 at the First Affiliated Hospital of Zhengzhou University. All the enrolled AF patients were hospitalized to undergo an initial RFCA procedure. Patients were excluded from the study if they met any of the following exclusion (I) had hypertrophic cardiomyopathy or advanced valvular heart disease; (II) had end-stage renal disease; (III) had thyroid dysfunction; (IV) had positive serum hepatitis B surface antigen or positive serum hepatitis C antibodies, or other chronic liver diseases, including autoimmune hepatitis, hereditary liver diseases, and drug-induced liver diseases; and/or (V) died or were lost to follow up.

This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethics Committee of The First Affiliated Hospital of Zhengzhou University (No. 2023-KY-0327), and the requirement of informed consent was waived due to the retrospective nature of the study.

Data collection and definitions

Standardized spreadsheets were created to systematically gather the retrospective data. The initial demographic and clinical variables included age, sex, current smoking and drinking status, diabetes mellitus (DM), hypertension, hyperlipidemia, coronary heart disease, a history of stroke/transient ischemic attack, heart failure, AF duration, and AF type (i.e., paroxysmal AF or persistent AF).

Blood samples were collected from the patients’ peripheral veins after fasting for more than 8 hours before ablation during hospitalization. Echocardiographic measurements of the left atrial (LA) diameter, left ventricular end-diastolic diameter, and left ventricular ejection fraction were recorded. Any medication use before ablation, including the use of anti-arrhythmic drugs (AADs), was documented. The data for calculating the CHA2DS2-VASc score were obtained from the hospital admission records (8). The duration of AF was calculated as the time from the date of initial symptom onset or first diagnosis of AF to the initial RFCA date. The diagnosis of NAFLD was based on the following three (I) ultrasound imaging of the liver revealing diffuse fatty infiltration, characterized by a generalized increase in near-field echogenicity, with an echo intensity greater than that of the spleen and kidneys, and a continuous decrease in far-field echogenicity; (II) the exclusion of other conditions that may lead to fatty liver disease, such as autoimmune liver diseases, drug-induced liver injury, and viral hepatitis; and (III) an absence of excessive alcohol consumption.

Ablation protocol

The RFCA procedure was performed as described previously (9,10). In brief, circumferential pulmonary vein (PV) isolation was performed in all patients using the CARTO system (Johnson & Johnson Medical, Biosense Webster, Inc., Irvine, CA, USA). Additional extra-PV ablations, such as isolation of the superior vena cava, roof line, tricuspid isthmus, mitral valve isthmus, and ablation of complex fractionated atrial electrograms, were carried out, particularly in patients with persistent AF. Patients with difficult mitral isthmus block underwent Marshall vein alcohol ablation therapy. If the AF rhythm persisted, synchronized biphasic electrical cardioversion was performed to restore sinus rhythm.

Outcomes and follow-up

The main study outcome was the recurrence of AF after ablation, assessed during a 1-year follow-up period. AF recurrence was defined as any atrial tachyarrhythmia lasting longer than 30 seconds as detected by electrocardiogram or Holter monitoring, after the initial 3-month blanking period. All the patients were prescribed AADs for three months post-ablation to prevent early recurrence. The subsequent use of AADs was determined through a collaborative decision-making process between physicians and patients.

All patient follow-up evaluations were conducted every three months during the first year after the procedure, either in outpatient settings or via telephone. (Some patients received outpatient treatment at local hospitals and were followed up by personnel via phone.) Each follow-up included a 12-lead electrocardiogram, 24-hour Holter monitoring, and clinical evaluation. Patients with AF-related symptoms were asked to schedule an extra outpatient visit.

Statistical analysis

The continuous variables are expressed as the mean ± standard deviation, or the median with the interquartile range, based on data distribution, and groups were compared using the Student’s t-test. The categorical data are presented as the count and percentage, and differences were assessed using the chi-square (χ^2^) test. The Kaplan-Meier method was used to plot cumulative AF recurrence curves following ablation, and differences between groups were evaluated using the log-rank test. Multivariable Cox proportional hazards models were then applied to investigate the association between risk factors and the recurrence of AF.

The initial model adjusted for age and gender, while the fully adjusted model further incorporated confounders from Model 1 alongside variables showing significant associations (P<0.05). Hazard ratios (HRs) and 95% confidence intervals (CIs) were reported based on the Cox regression outcomes. To evaluate the added predictive value of including NAFLD in the baseline risk model for AF recurrence, metrics such as the C-statistic, continuous net reclassification improvement (NRI), and integrated discrimination improvement (IDI) were calculated. Additionally, subgroup analyses were performed to determine whether the predictive effect of NAFLD remained consistent across various demographic and comorbidity subgroups. The statistical analyses were performed using IBM SPSS 26.0 and R 3.6.1. All the tests were two-sided. A P value <0.05 was considered statistically significant.

Results

Baseline characteristics

A total of 1,413 AF patients who underwent successful RFCA were initially screened, of whom 1,182 were included in the final analysis (Figure S1). Over a 1-year follow-up period, AF recurrence was observed in 356 patients (30.1%). The baseline characteristics of the patients with and without AF recurrence are presented in Table 1. Within 3 months of ablation, 1,088 (92.0%) patients took AADs, of whom 846 took amiodarone, 199 took propafenone, and 43 took betalecker. There was no significant difference between the recurrence group and the non-recurrence group. Compared to the patients without AF recurrence, those with AF recurrence showed higher rates of current alcohol use, DM, persistent AF, and an AF duration of 24 months or more. They also had higher body mass index (BMI), CHA2DS2-VASc, and LA diameter values (P<0.05). Notably, the prevalence of NAFLD was significantly higher in patients with AF recurrence than in those without AF recurrence.

The predictive value of NAFLD for AF recurrence

The Kaplan-Meier curves showed that the 1-year cumulative incidence of AF recurrence was significantly higher in the patients with NAFLD than in those without (Figure 1). Further, we examined the association between NAFLD and the recurrence of AF after ablation by adjusting for different risk factors. Model 1 was adjusted for age and gender, while Model 2 was adjusted for the risk factors identified in the univariate analysis. Multivariate Cox regression showed that NAFLD remained an independent risk factor for the recurrence of AF after controlling for model 2 (HR =1.37, 95% CI: 1.10–1.70, P=0.005; Table 2).

NAFLD incremental predictive value for AF recurrence

Table 3 shows the predictive value of NAFLD for AF recurrence. Adding NAFLD to the baseline risk model significantly enhanced its predictive accuracy, with the C-statistic increasing from 0.672 to 0.686 (P=0.03). Moreover, significant improvements were observed in both the NRI and IDI values after the inclusion of NAFLD in the model (P<0.05).

The predictive value of NAFLD for AF recurrence in various subgroups

Multiple subgroup analyses were performed to evaluate whether the predictive significance of NAFLD was consistent across different demographic groups and comorbid conditions. Figure 2 shows the association between NAFLD and the recurrence of AF, stratified by age, sex, BMI, DM, hypertension, hyperlipidemia, AF duration, and AF type. NAFLD was found to be a significant predictor of AF recurrence across various subgroups. Further, the findings revealed a significant interaction between sex and NAFLD in relation to the risk of AF recurrence (P value for interaction =0.02). Notably, a significant interaction was also found between NAFLD and DM in relation to AF recurrence after RFCA (P value for interaction =0.049).

Association between NAFLD and AF recurrence based on DM or sex status

Quadripartite stratification by NAFLD and DM/female factors identified a high-risk the NAFLD patients who had DM or were female had a significantly higher incidence of AF recurrence than the other patients (Figure S2A,S2B). After adjusting for different risk factors, the factors of DM or female sex were both found to be independently associated with AF recurrence in the NAFLD patients (Table S1).

Discussion

This study found that NAFLD was significantly associated with an increased risk of AF recurrence after RFCA, independent of traditional risk factors. The observed associations remained statistically significant across all adjusted models after controlling for conventional cardiovascular risk factors. The combination of NAFLD with the factors of DM or female sex resulted in a particularly elevated risk of recurrence. Adding NAFLD to baseline prediction models enhanced their risk stratification accuracy, highlighting its potential clinical value. These results address a critical knowledge gap in AF management, offering an evidence-based rationale for incorporating NAFLD status into risk-stratified follow-up protocols after ablation.

There has been a marked increase in the use of catheter ablation for AF in clinical practice worldwide, reflecting its evolving role in rhythm control strategies. In our study, the patients with recurrent AF had a higher prevalence of DM, a higher BMI, and a longer history of AF. The recurrent patients also had more complications than those without recurrence (7). Lifestyle changes, such as dietary alterations, physical activity and weight control, may reduce the recurrence of AF after ablation (11). Individuals with AF, particularly those with NAFLD, maintain a heightened susceptibility to an increased risk of AF (5).

NAFLD is considered a hepatic manifestation of metabolic syndrome, occurring in 60–90% of patients with DM or obesity (12). Mohanty* et al. *(13) reported that metabolic syndrome significantly influences arrhythmia recurrence after ablation, with patients affected by metabolic syndrome exhibiting significantly higher recurrence rates than controls. Additionally, a systematic review of 13 cohort studies reported a modest but statistically significant association between NAFLD and an increased risk of AF (HR =1.18; 95% CI: 1.12–1.23) (14). Patients with metabolic dysfunction-associated fatty liver disease have also been reported to have a significantly higher rate of incident AF (HR =1.26, 95% CI: 1.18–1.35) (15). NAFLD predisposes patients to AF, regardless of the known risk factors for atherosclerosis (16). Another study found that the risk of AF increased with the severity of NAFLD across all age groups. These findings indicate an independent association between NAFLD and AF, regardless of age, except in patients aged ≥70 years (5). Another study found that NAFLD was associated with a higher 1-year risk of adverse events in AF patients (17). In the current era, where there is a greater emphasis on the critical role of pre-ablation risk factor management, NAFLD presents an additional modifiable target for optimizing medical management before ablation.

In the present study, we found that NAFLD independently predicted AF recurrence after ablation. Further, we demonstrated improved prediction accuracy for AF recurrence when NAFLD was added to the baseline risk model. NAFLD exhibited a significant interaction with AF recurrence in both the DM and female patients. DM has traditionally served as a clinical indicator of a greater AF burden and more advanced atrial remodeling, indicating worse clinical outcomes (18,19). A higher burden of comorbidities, especially DM, increases the risk of AF. DM may be a sign of increased metabolic disorders in patients with NAFLD (20). Our findings suggest that the co-existence of NAFLD and DM may synergistically exacerbate metabolic dysregulation, implying that targeted metabolic interventions could potentially attenuate late AF recurrence (21,22).

Additionally, female AF patients have been reported to exhibit a higher frequency of extra-PV triggers and lower LA voltage (23). LA volume was found to be an independent factor for AF recurrence after ablation. Previous studies have shown that LA volume is generally larger in females than in males. Histopathological data reveal sex-specific patterns of collagen deposition, with females exhibiting more diffuse interstitial fibrosis (24,25). The presence of diffuse fibrosis in females may lead to the formation of complex regions that act as electrical conduction barriers, thereby necessitating broader ablation coverage to effectively target latent reentrant pathways (26,27). With advancing age, women exhibit progressive elevation in central aortic pressure, which promotes the development of left ventricular hypertrophy and diastolic dysfunction, collectively creating a pathophysiological substrate highly susceptible to AF initiation and maintenance (28). Researching diverse groups from multiple perspectives can extend our understanding of NAFLD and help identify the most clinically useful factors. Further studies are urgently needed to confirm these findings.

Presently, the precise pathophysiological mechanisms underlying the NAFLD-AF association require further elucidation. Current evidence suggests some predominant pathways. Systemic inflammation and oxidative stress may be significant factors in the development of AF in patients with NAFLD (29). Adiponectin is a key adipokine produced predominantly by adipocytes and plays a crucial role in maintaining metabolism. It may serve as a bridge between adipose tissue, cardiomyocytes, and arrhythmias (30,31). NAFLD is associated with an increase in adipose tissue (32). Emerging evidence suggests that ectopic fat deposition functions as a metabolically active endocrine organ, secreting pro-inflammatory cytokines and vasoactive mediators that directly mediate myocardial architectural remodeling and electromechanical dysfunction (33). Research indicates that NAFLD is associated with cardiac autonomic dysfunction, which may be related to the occurrence and progression of AF. The association between NAFLD and cardiac remodeling, as well as arrhythmias, may be mediated by shared pathogenic mechanisms, such as metabolic dysregulation and systemic insulin resistance, which are central features of NAFLD pathophysiology (34).

This study had several limitations. First, despite the implementation of standardized inclusion criteria, the retrospective design carries inherent risks of selection bias. The study is also limited by the potential for unmeasured confounding variables. Indeed, when using multivariate regression to analyze the recurrence factors of AF, the actual date of AF recurrence might not have been the same as the date of the contact, which might have led to deviations in the analysis results. Second, the LA assessment relied on diameter measurements rather than the guideline-endorsed volumetric analysis, which might have compromised the anatomical evaluation accuracy. This study did not include parameters such as LA volume index and global longitudinal strain. Third, despite patients being encouraged to report any suspected symptoms, asymptomatic arrhythmias might have led to missed AF recurrences. AF recurrence monitoring depended on symptomatic reporting coupled with routine electrocardiogram/Holter protocols, and advanced detection tools were not deployed, increasing the likelihood of missing subclinical arrhythmic events. The multivariate adjustments addressed documented confounders; however, residual confounding from unmeasured variables might persist in the observed associations. Additionally, the exclusion of patients who died during follow-up may have affected the results, and selection bias could not be avoided. The therapeutic implications of NAFLD management for post-ablation outcomes require validation through prospective randomized trials, particularly those targeting the cardio-hepatic axis mechanisms.

Conclusions

NAFLD was independently associated with the recurrence of AF following ablation. NAFLD may serve as a valuable tool for stratifying the risk of AF recurrence following ablation. The combination of NAFLD with the factors of either DM or female sex was associated with a particularly poor prognosis. Our findings show that NAFLD may serve as a dependable marker for assessing AF recurrence risk in clinical practice.

Supplementary