Long-term cumulative and trajectory patterns of the C-reactive protein–triglyceride–glucose index and cardiometabolic multimorbidity: the mediating effects of adiposity indices

Background Cardiometabolic multimorbidity (CMM) is increasingly prevalent in ageing populations and confers substantial clinical and societal burdens. The C-reactive protein–triglyceride–glucose index (CTI) integrates inflammatory and metabolic information, but evidence linking long-term CTI patterns to incident CMM and the role of adiposity as a potential mediator remains limited. Methods We included 5,778 participants aged ≥ 45 years who were free of CMM at baseline. Exposures were baseline CTI in 2012, cumulative CTI(cuCTI) from 2012 to 2015, and CTI change-pattern (trajectory) groups derived from repeated measurements. Incident CMM was assessed in 2018. Multivariable logistic regression models were pooled across multiple imputations; restricted cubic splines examined dose–response. Prediction performance was evaluated using C-statistics (DeLong test), net reclassification improvement, and integrated discrimination improvement. Mediation by adiposity indices was assessed. Results Incident CMM occurred in 333 participants (5.8%). In fully adjusted models, higher baseline CTI and greater cumulative CTI exposure were independently associated with higher odds of incident CMM (baseline CTI per IQR: OR 1.545 (95% CI 1.344–1.776); cuCTI per IQR: OR 1.651 (95% CI 1.429–1.907), both P < 0.001). Change-pattern (trajectory) analyses further indicated substantially higher risk among participants with persistently elevated CTI compared with those with persistently low CTI (OR 3.078 (95% CI 2.169–4.369), P < 0.001). Adding baseline CTI, cuCTI, or change-pattern grouping improved discrimination beyond the basic model and improved reclassification. Adiposity indices partially mediated the CTI–CMM associations. Conclusions Higher cumulative CTI and persistently elevated CTI change patterns were associated with incident cardiometabolic multimorbidity and improved risk prediction, with partial mediation by adiposity.