Insulin resistance as a predictor of long-term adverse cardiovascular event risk in patients with atrial fibrillation following radiofrequency catheter ablation.

Background: Catheter ablation is the first-line therapy for atrial fibrillation (AF), yet a significant proportion of patients experience major adverse cardiovascular events (MACEs) post-ablation, highlighting a need for improved risk stratification. Objective: This study aimed to evaluate the association between four non-insulin-based insulin resistance (IR) indices and long-term MACEs in AF patients undergoing radiofrequency catheter ablation (RFCA), and to determine their incremental predictive value beyond established clinical risk factors. Methods: A retrospective observational study was conducted on 922 non-valvular AF patients who underwent index RFCA between March 2017 and July 2023. The primary endpoint was a composite of MACEs, defined as all-cause death, late AF recurrence, heart failure events, and stroke occurring after the 3-month blanking period. Associations between the four IR indices and MACEs were examined using multivariable Cox proportional hazards regression, restricted cubic spline analysis, and Kaplan-Meier survival estimation. The incremental predictive value of each index over a baseline clinical model was quantified using the C-statistic, net reclassification improvement (NRI), and integrated discrimination improvement (IDI). Results: Over a median follow-up of 36 months, MACEs occurred in 242 patients (26.25%). After full multivariable adjustment, METS-IR and TyG-BMI were independently and significantly associated with elevated MACE risk across all models (both P < 0.001), while the TG/HDL-C ratio and TyG index did not reach statistical significance in fully adjusted models. METS-IR demonstrated an approximately linear dose-response relationship with MACEs on restricted cubic spline analysis. When added to the baseline clinical model (C-statistic: 0.702), METS-IR conferred the greatest incremental predictive gain, significantly improving the C-statistic to 0.717 (P < 0.001), alongside meaningful NRI (0.291, 95% CI: 0.126-0.426, P < 0.001) and IDI (0.015, 95% CI: 0.004-0.035, P = 0.004). Conclusion: Among the evaluated non-insulin-based IR indices, METS-IR demonstrated a statistically significant, albeit modest, incremental improvement in the prediction of composite MACE in AF patients following RFCA. These findings support the use of METS-IR in post-ablation risk assessment and may contribute to improved patient outcomes.