The association between residual cholesterol to high-density lipoprotein cholesterol ratio and type 2 diabetes risk in elderly populations

Purpose The remnant cholesterol to high-density lipoprotein cholesterol ratio (RC/HDL-C) is a novel biomarker of metabolic disturbances. Prior studies have indicated a close association between elevated RC/HDL-C and the occurrence and progression of cardiovascular, cerebrovascular, and metabolic diseases. However, longitudinal data on the relationship between RC/HDL-C and type 2 diabetes (T2D) risk remain limited, particularly among high-risk elderly populations. This study aims to clarify the association between RC/HDL-C and incident T2D in Chinese adults, offering evidence for early prevention and detection of T2D. Patients and methods A follow-up study was conducted in Kunshan, China, from January 2018 to July 2021, involving 7655 participants aged 60 years or older. Cox proportional hazards models were used to evaluate the independent effect of the RC/HDL-C level on the risk of T2D. The relationship between RC/HDL-C and T2D risk was visually demonstrated using the Kaplan-Meier method and restricted cubic splines (RCS). Results During a median follow-up of 3.87 years, 783 participants (10.23%) developed T2D. A fully adjusted Cox proportional hazards model showed a positive and independent association between RC/HDL-C and T2D risk (HR = 1.12, 95% CI: 1.06-1.17, P < 0.001). The highest RC/HDL-C quartile (Q4) had the highest cumulative incidence of T2D (Log-rank test, P < 0.001). Furthermore, RCS analysis revealed a non-linear relationship between RC/HDL-C and T2D risk. Receiver operating characteristic (ROC) curve analysis revealed that the RC/HDL-C ratio exhibited the largest area under the curve (AUC = 0.601, 95% CI: 0.580–0.623), indicating modest but statistically significant predictive ability. Subgroup analysis further validated the robustness of these results. Conclusion We found that older adults with elevated RC/HDL-C levels have a higher risk of incident T2D. RC/HDL-C is an independent predictor of incident T2D in the elderly and may serve as a valuable adjunct to enhance risk stratification within existing prediction models.