Cardiovasc Diabetol. 2026 Feb 3. doi: 10.1186/s12933-025-03043-6. Online ahead of print.
ABSTRACT
BACKGROUND: The atherogenic index of plasma (AIP) serves as a crucial indicator for assessing atherosclerotic risk. It reflects the degree of dyslipidaemia and cardiovascular disease (CVD) risk. The cardiometabolic index (CMI) provides a comprehensive evaluation of obesity-related metabolic risk, acting as a key biomarker for predicting multiple cardiometabolic diseases. The relationship between AIP and CMI in patients with non-alcoholic fatty liver disease (NAFLD) and mortality or CVD risk remains unclear.
METHODS: This study included 5792 adult (≥ 18 years) NAFLD patients from the US National Health and Nutrition Examination Survey (NHANES, 1999-2018). Weighted logistic regression and Cox proportional hazards models were employed to investigate the association between AIP, CMI and all-cause mortality, CVD mortality and CVD risk. Restricted cubic spline (RCS) curves assessed non-linear associations. Subgroup analyses and mediation analyses examined the effect modifiers and mediators. The incremental predictive value of AIP and CMI was evaluated. Sensitivity analyses were conducted to validate the robustness.
RESULTS: During follow-up, 721 all-cause deaths (including 241 CVD deaths) and 726 total CVD events were recorded. After adjusting for confounding factors, patients in the highest quartiles of AIP and CMI had a significantly higher risk of specific CVD events. The strongest association was observed for CHF (AIP: OR = 3.157, 95% CI 1.684, 5.922, p < 0.001; CMI: OR = 3.604, 95% CI 1.843, 7.047, p < 0.001), followed by heart attack and CHD. CMI consistently demonstrated a stronger effect than AIP. RCS analysis indicates a non-linear relationship between CMI and CHD, angina pectoris. Subgroup analysis revealed that both AIP and CMI demonstrated high predictive value for all-cause mortality in the 40-60 age cohort. Mediation analysis revealed that Mets, NLR, hypertension and HOMA-IR partially mediated the aforementioned associations. The inclusion of AIP and CMI partially improved the predictive capability of the basic model. Sensitivity analyses validated the robustness of these findings.
CONCLUSIONS: In patients with NAFLD, CMI is a stronger predictor of non-fatal CVD than AIP. While both indices show limited value for predicting mortality, CMI holds promise as a practical supplementary tool for clinical risk assessment.
PMID:41634708 | DOI:10.1186/s12933-025-03043-6
