The association of the triglyceride-glucose index with the risk of atrial fibrillation: Analysis of the UK Biobank

Background and aims: The relationship between the triglyceride-glucose (TyG) index and the incidence of atrial fibrillation (AF) remains insufficiently explored. This investigation aims to elucidate the association between the TyG index and the long-term risk of developing AF.

Methods and results: This cohort study analyzed data from 409,705 participants sourced from the UK Biobank database. Participants were stratified into three groups based on TyG index tertiles. The association between the TyG index and AF was evaluated using Cox proportional hazards models. Restricted cubic spline (RCS) analysis was employed to investigate potential linear or nonlinear relationships. During a mean follow-up period of 13.9 years, 26,092 AF cases were recorded. Compared with the T2 group, participants in the T1 group and T3 group presented a significantly higher risk of AF (T1: HR: 1.22, 95%CI: 1.17-1.27; T3: HR: 1.09, 95%CI: 1.05-1.14). RCS analysis documented a U-shaped relationship between the TyG index and the risk of AF (P for non-linearity <0.001). In non-type 2 diabetes (T2D) participants, TyG levels were associated with AF risk in a U-shaped relationship. Among T2D participants, only the T3 group had an increased risk of AF (reverse "L" pattern). The U-shaped relationship between TyG levels and AF risk remained consistent across heart valve disease (HVD) and non-HVD patients, as well as different strata of genetic susceptibility to AF.

Conclusions: This study demonstrates a U-shaped association between the TyG index and the risks of AF, underscoring the index's potential utility in identifying individuals at elevated risk for these conditions.