Triglyceride-glucose-related metabolic indices, phenotypic aging, and incident pulmonary embolism

TyG-related metabolic indices are practical markers of metabolic dysfunction, but prospective evidence linking these indices to incident pulmonary embolism (PE), phenotypic aging, and genetic susceptibility remains limited.

Among 345,620 UK Biobank participants enrolled between 2006 and 2010 and followed for a median of 13.6 years through October 31, 2022, we examined five baseline TyG-related metabolic surrogate indices in relation to incident PE identified from linked health records. Cox models, restricted cubic splines, competing-risk sensitivity analyses, exploratory PhenoAgeAccel-related decomposition, and exploratory interaction analyses involving the PE genetic susceptibility score were performed.

During follow-up, 6206 incident PE cases occurred. In Cox models, all five TyG-related metabolic indices were positively associated with incident PE when expressed per 1-SD increase. The associations were directionally consistent across competing-risk, landmark, imputed-data, and PhenoAgeAccel-excluding sensitivity analyses. PhenoAgeAccel-related decomposition suggested statistical attenuation but was interpreted as exploratory rather than causal mediation. Multiplicative interactions with PE genetic susceptibility score group were not statistically significant after FDR correction, whereas exploratory additive-scale interaction signals were observed for several indices.

Higher TyG-related metabolic indices were associated with incident PE in this large prospective cohort. PhenoAgeAccel-related decomposition and genetic susceptibility analyses were exploratory and require external replication.