Abstract
<title>Abstract</title> <p>Background Cardiovascular-kidney-metabolic abnormalities may precede overt diabetes, but practical markers of longitudinal cross-system deterioration remain limited. We conducted a longitudinal cohort analysis in the China Health and Retirement Longitudinal Study, using 2015 as the index wave, to examine whether change in the cystatin C-to-high-density lipoprotein cholesterol ratio (CysC/HDL-C) from 2011 to 2015 was associated with subsequent survey-defined heart disease among non-diabetic adults with early cardiovascular-kidney-metabolic abnormalities. Results The analysis included 3,133 participants; 273 reported newly diagnosed survey-defined heart disease during a median post-2015 follow-up interval of 3 years. Event rates increased from 7.0% in the lowest quartile to 10.1% in the highest quartile of Δln(CysC/HDL-C). Each 1-SD increase in Δln(CysC/HDL-C) was associated with higher risk in the primary adjusted model (hazard ratio [HR] 1.18, 95% confidence interval [CI] 1.03–1.36; P = 0.017) and after additional adjustment for estimated glomerular filtration rate, C-reactive protein, glycated hemoglobin, and fasting glucose (HR 1.21, 95% CI 1.05–1.40; P = 0.010). Interaction tests by cardiovascular-kidney-metabolic stage and sex were not significant, and the increment in model discrimination was modest. Conclusions Longitudinal worsening of CysC/HDL-C was modestly associated with subsequent survey-defined heart disease in non-diabetic adults with early cardiovascular-kidney-metabolic abnormalities. This finding may provide adjunctive epidemiological information but does not support use of the ratio as a stand-alone clinical prediction tool.</p>