Abstract
BACKGROUND: With the rising proportion of recipients with type 2 diabetes (T2D) undergoing simultaneous pancreas-kidney transplantation (SPK), cardiovascular complications remain the leading cause of post-transplant mortality. However, tools for the early prediction of cardiovascular risk are lacking. This study evaluated the predictive value of longitudinal triglyceride-glucose (TyG) index trajectories for long-term major adverse cardiovascular diseases events (MACE) after SPK. METHODS: In this retrospective single-center study, 106 patients with T2D who underwent SPK were analyzed. Latent class mixed modeling was applied to categorize TyG index trajectories across four time points (pre-transplant baseline, 3/6/12 months post-transplant). Associations between trajectory patterns and MACE were assessed using Cox regression analysis, and model performance was validated using optimism-corrected concordance indices. RESULTS: Two distinct groups were identified, a metabolic improvement group (72.6%) with high baseline TyG and sustained post-transplant reduction, and a metabolic worsening group (27.4%) with low baseline TyG and progressive elevation. Over a median follow-up of 5.68 years, the metabolic worsening group exhibited a significantly higher MACE incidence (24.1% vs. 7.8%, P = 0.041), an association that remained significant after adjustment for confounders (adjusted hazard ratio [HR] = 3.52; 95% confidence interval [CI]: 1.17-10.6; P = 0.025). Furthermore, the metabolic worsening trajectory independently predicted reduced kidney graft survival (adjusted HR = 3.35; 95% CI: 1.04-10.8; P = 0.043). Pre-transplant cardiovascular history also was as a significant predictor of MACE risk (adjusted HR = 3.57; 95% CI: 1.15-11.1; P = 0.028). The predictive model incorporating these factors demonstrated robust predictive accuracy, with an optimism-corrected C-index of 0.741. CONCLUSIONS: Serial TyG index monitoring identified dynamic post-SPK metabolic risk patterns and distinguished high-risk subgroups for targeted interventions. Integrating TyG trajectories with clinical predictors enhances MACE risk stratification, thus offering a pragmatic tool for personalized cardiovascular prevention in T2D transplant recipients. These findings also suggest the potential utility of TyG trajectories in predicting kidney graft outcomes.