We enrolled 478 patients who underwent the first radiofrequency catheter ablation for AF between January 2021 and March 2024 at the First People’s Hospital of Yunnan Province (Kunming, China). The exclusion criteria were as follows: 1) moderate-to-severe valvular disease; 2) uncontrolled thyroid dysfunction; 3) left atrial thrombosis; 4) acute coronary syndrome, myocardial infarction, and cardiac surgery in the previous 3 months; 5) contraindication of anticoagulation; 6) pregnancy; 7) hepatic or renal failure; and 8) patients who died or who were lost to follow-up. The study is in compliance with the principles of the Declaration of Helsinki and was approved by the Ethics Committee of Yunnan First People’s Hospital.

Clinical data including age, sex, BMI, waist circumference, hypertension (HT), diabetes mellitus (DM), history of stroke, heart failure (HF), and type of AF (paroxysmal AF or persistent AF) were collected. The CHA₂DS₂-VASc score was calculated for each patient. Serum blood biomarkers such as fasting plasma glucose (fGLU), creatinine, total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), and high-density lipoprotein cholesterol (HDL-C) were determined. Transthoracic echocardiography was employed to record the left atrial diameter (LAD) and the left ventricular ejection fraction (EF). The VAT surrogates were calculated as follows: BMI = weight (kg)/height² (m²); WHtR = waist (cm)/height (cm); LAP(women) = TG (mmol/L) * [WC (cm) − 58]; LAP(men) = TG (mmol/L) * [WC (cm) − 65]; CMI = TG (mmol/L)/HDL-C (mmol/L) * WHtR; METS-IR = Ln[2 × FPG (mg/dl) + TG (mg/dl)] × BMI (kg/m²)/Ln [HDL-C (mg/dl); and METS-VF = 4.466 + 0.011 × [Ln(METS-VF)]³ + 3.239 × [Ln(WHtR)]³ + 0.319 × Sex (men = 1, women= 0) + 0.594 × [Ln(Age) (years)] (10).

The details regarding electrophysiology and catheter ablation are available in our previous study (11). In brief, an open irrigated catheter (ST Catheter, Biosense-Webster, Diamond Bar, CA, USA) was used to perform circumferential pulmonary vein (PV) isolation. Additional ablations in the superior vena cava or other non-PV triggers were performed when mappable AF triggers were available.

Patients were followed up at the outpatient department at 3, 6, 9, and 12 months after ablation and every 6 months thereafter, for a medium follow-up duration of 18 ± 9.6 months. At each visit, a 12-lead electrocardiogram and 24-h Holter monitoring were performed. When a patient showed symptoms of palpitations, the Holter data were obtained to evaluate arrhythmia. AF recurrence was defined as any episode of AF or atrial tachycardia lasting more than 30 s.

Descriptive statistics are reported as frequencies and percentages for categorical variables and as medians with interquartile ranges (Q1–Q3) for continuous variables. Continuous variables were compared using the Wilcoxon rank-sum test or the Kruskal–Wallis test, while categorical variables were compared using the chi-square test. Kaplan–Meier survival curves were generated to visualize the time to AF recurrence stratified by the METS-VF quartiles, and differences between groups were assessed using the log-rank test. The association between the clinical/metabolic variables and AF recurrence was assessed using Cox proportional hazards regression. Univariable Cox models were performed first, followed by multivariable models adjusting for potential confounders. The final multivariate model included age, sex, BMI, stroke, HT, HF, cardiovascular disease (CVD), DM, TC, LDL-C, HDL-C, triglycerides (Tg), fGLU, and LAD. Restricted cubic spline (RCS) modeling was used to explore the dose–response relationship and the potential non-linearity between METS-VF and AF recurrence risk. Time-dependent ROC curve analysis was conducted at 12, 24, and 36 months to evaluate and compare the predictive performance of METS-VF, CMI, LAP, and METS-IR. The area under the curve (AUC) values were compared using DeLong’s test. All data analyses were performed using R or SPSS. A p < 0.05 was considered statistically significant.

This study included 478 patients (age, 58.9 ± 10.6 years; 62.8% men) with either paroxysmal (n = 370) or persistent AF (n = 108). The demographic characteristics, the clinical and laboratory data, and the VAT surrogates are summarized in Table 1 . After a mean follow-up period of 18.0 ± 9.6 months, AF recurrence was observed in 112 (23.4%) patients. As shown in Table 1 , patients with AF recurrence exhibited larger LAD (39.4 ± 5.7 vs. 37.4 ± 5.9, p = 0.002), higher CMI (0.68 vs. 0.56, p = 0.039), higher LAP (28.1 vs. 23.5, p = 0.023), higher MET-IR (34.29 ± 7.68 vs. 32.52 ± 6.94, p = 0.02), and higher METS-VF (6.38 ± 0.50 vs. 5.99 ± 0.53, p = 0.004) compared to patients without AF recurrence. Furthermore, patients were stratified into four groups according to the METS-VF quartiles ( Table 2 ). The quartile thresholds for METS-VF were determined as 5.90, 6.35, and 6.68. Patients in the highest quartile of METS-VF had higher BMI, larger LAD, and higher fGLU, CMI, LAP, and METS-IR than those in the other three groups (all p < 0.0001).

Kaplan–Meier analysis was performed to estimate the time to AF recurrence after catheter ablation across the different METS_VF risk strata ( Figure 1 ). Kaplan–Meier analysis revealed a significant difference in AF recurrence across the METS-VF quartiles (log-rank p = 0.021), with higher METS-VF scores associated with an increased risk of recurrence following catheter ablation. Cox proportional hazards regression models were then utilized to identify whether METS-VF and other metabolic indices were independently associated with AF recurrence over time. In the univariable Cox regression analysis ( Table 3 ), several clinical and metabolic variables were assessed for their association with AF recurrence following catheter ablation. Among them, LAD and a higher METS-VF were independently associated with the recurrence of AF after ablation. Multivariable Cox regression ( Table 3 ) confirmed METS-VF as an independent predictor of AF recurrence after adjusting for confounders. To further evaluate risk stratification, METS-VF was analyzed as a categorical variable using quartiles ( Table 4 ). In all three adjustment models, patients in the highest quartile (Q4) showed a significantly elevated risk of AF recurrence compared with those in Q1.

To evaluate the discriminative ability of METS-VF in predicting AF recurrence after catheter ablation, time-dependent ROC curve analyses were conducted and the results compared against three established metabolic indices: CMI, LAP, and METS-IR. Figure 2 shows the time-dependent ROC curves generated at 12, 24, and 36 months of follow-up. The corresponding AUC values are presented in Table 5 . Comparisons were performed using DeLong’s test to assess statistical significance. At the 12-month mark, METS-VF demonstrated significantly better predictive performance than LAP (p = 0.027), CMI (p = 0.003), and METS-IR (p = 0.013). However, the differences in the AUCs between METS-VF and the comparator models at 24 and 36 months were not statistically significant (all p > 0.3).

The association between METS-VF and AF recurrence was further examined using RCS analysis. As shown in Figure 3 , a linear and positive dose–response relationship was observed between METS-VF and the risk of AF recurrence. The overall association was statistically significant (p = 0.02); however, no evidence of non-linearity was detected (p = 0.62). This suggests that higher METS-VF scores are linearly associated with an increased risk of AF recurrence following catheter ablation.

This study has several limitations. Firstly, this is a retrospective study based on patients who were referred for AF ablation in a single center, which needs validation in other populations. Secondly, the mechanisms between VAT and AF recurrence were not fully understood, and more basic research is needed to investigate these.