Joint association of five insulin resistance indices and frailty index with incident cardiometabolic multimorbidity in individuals with cardiovascular–kidney–metabolic syndrome stages 0–3: a prospective cohort study

@article{zou2026joint,
  title = {{Joint association of five insulin resistance indices and frailty index with incident cardiometabolic multimorbidity in individuals with cardiovascular–kidney–metabolic syndrome stages 0–3: a prospective cohort study}},
  author = {Jiawang Zou and Hailan Xiong and Shuchang Ding and Peishan Wen and Yushan Lei and Bo Liu and Jing Zhou and Yanhua Wen and Jiangbo Xie},
  journal = {Cardiovascular Diabetology},
  year = {2026},
  month = sep,
  doi = {10.1186/s12933-026-03364-0},
  url = {https://doi.org/10.1186/s12933-026-03364-0},
  note = {CC BY-NC-ND 4.0},
}

TY  - JOUR
TI  - Joint association of five insulin resistance indices and frailty index with incident cardiometabolic multimorbidity in individuals with cardiovascular–kidney–metabolic syndrome stages 0–3: a prospective cohort study
AU  - Jiawang Zou
AU  - Hailan Xiong
AU  - Shuchang Ding
AU  - Peishan Wen
AU  - Yushan Lei
AU  - Bo Liu
AU  - Jing Zhou
AU  - Yanhua Wen
AU  - Jiangbo Xie
PY  - 2026
DA  - 2026/09/06
JO  - Cardiovascular Diabetology
DO  - 10.1186/s12933-026-03364-0
UR  - https://doi.org/10.1186/s12933-026-03364-0
AB  - Background: Insulin resistance and frailty are closely related to cardiometabolic deterioration, yet their joint contribution to the development of cardiometabolic multimorbidity (CMM) among individuals with early-to-intermediate cardiovascular–kidney–metabolic (CKM) syndrome remains unclear. This study examined the associations of five insulin resistance-frailty composite indices and the joint status of insulin resistance indices and frailty index with incident CMM in Chinese middle-aged and older adults with CKM stages 0–3.  Methods: This prospective cohort study included 6023 participants from the China Health and Retirement Longitudinal Study who had CKM stages 0–3 and were free of CMM at baseline. Insulin resistance was assessed using the triglyceride–glucose index (TyG), C-reactive protein-triglyceride glucose index (CTI), cholesterol, high-density lipoprotein, and glucose index (CHG), stress hyperglycemia ratio (SHR), and estimated glucose disposal rate (eGDR). Frailty was evaluated using a 32-item frailty index (FI). Composite indices included TyG-FI, CTI-FI, CHG-FI, SHR-FI, and eGDR/FI. Participants were also cross-classified into joint exposure groups according to baseline median-derived cutoff values of each insulin resistance index and FI. Cox proportional hazards models were used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs). Restricted cubic spline, subgroup, additive interaction, ROC, decision curve, NRI, IDI, and multiple-imputation sensitivity analyses were performed.  Results: During a median follow-up of 9.0 years, 399 participants developed incident CMM. In fully adjusted models, each 1-SD increment in TyG-FI, CTI-FI, CHG-FI, and SHR-FI was associated with a 29%, 30%, 31%, and 27% higher risk of incident CMM, respectively (TyG-FI: HR 1.29, 95% CI 1.18–1.41; CTI-FI: HR 1.30, 95% CI 1.19–1.41; CHG-FI: HR 1.31, 95% CI 1.20–1.43; SHR-FI: HR 1.27, 95% CI 1.17–1.39). Conversely, each 1-SD increment in eGDR/FI was associated with a 39% lower CMM risk (HR 0.61, 95% CI 0.48–0.78). Restricted cubic spline analyses showed significant nonlinear dose-response relationships. The threshold points were 1.18 for TyG-FI, 1.17 for CTI-FI, 0.87 for CHG-FI, 0.13 for SHR-FI, and 337.74 for eGDR/FI; risks increased with higher TyG-FI, CTI-FI, CHG-FI, and SHR-FI, particularly before the thresholds, whereas CMM risk declined as eGDR/FI increased and then tended to plateau. Compared with the lowest quartile, the highest quartiles of TyG-FI, CTI-FI, CHG-FI, and SHR-FI were associated with markedly higher risks of incident CMM, whereas the highest quartile of eGDR/FI was associated with lower risk. Joint analyses showed that participants with both high insulin resistance indices and high FI had the greatest CMM risk, particularly for CHG and FI (HR 6.47, 95% CI 4.15–10.10) and CTI and FI (HR 6.27, 95% CI 4.01–9.81). Adding FI increased the AUCs of the TyG, CTI, CHG, SHR, and eGDR models from 0.668 to 0.708, 0.683 to 0.718, 0.709 to 0.742, 0.547 to 0.654, and 0.702 to 0.723, respectively. NRI estimates ranged from 3.89 to 16.01% and IDI estimates from 0.73 to 1.40%, with all corresponding P values < 0.01; decision curve analyses generally showed greater or comparable standardized net benefit for the IR-plus-FI models. Additive interaction analysis showed a positive additive interaction between CHG and FI, with a relative excess risk due to interaction of 1.47 (95% CI 0.08–2.87) and an attributable proportion of 0.23 (95% CI 0.03–0.42).  Conclusion: Among Chinese middle-aged and older adults with CKM stages 0–3, insulin resistance-frailty composite indices were independently associated with incident CMM. Joint analyses revealed that individuals with concurrent insulin resistance and greater frailty exhibited the highest risk, suggesting that metabolic dysfunction combined with frailty constitutes a high‑risk CKM phenotype predisposed to progression toward CMM. Adding FI improved the exploratory discrimination and decision-analytic performance of individual insulin-resistance indices. These findings support the integrated assessment of insulin resistance and frailty for early‑stage risk stratification and prevention of CMM among individuals within the CKM spectrum.  Graphical abstract:  Supplementary Information: The online version contains supplementary material available at https://doi.org/10.1186/s12933-026-03364-0.
SN  - 1475-2840
ER  -