Remnant cholesterol variability and diabetes mortality risk in Chinese older adults: a multi-city cohort study

Remnant cholesterol (RC) is a known risk factor for cardiovascular disease and is associated with incident type 2 diabetes. Beyond single measurements, variability in RC may reflect additional dimensions of metabolic instability. However, the association between RC variability and the risk of diabetes mortality, particularly in older adults, remains unclear. This was an observational study including eligible participants aged ≥ 65 years from the Multi-city Elderly Health Examination Cohort Study who underwent three health examinations from 2017 to 2020. RC variability was assessed using standard deviation (SD), coefficient of variation (CV), average real variability (ARV) and variability independent of the mean (VIM). Cox models were used to evaluate the associations of RC variability and RC change with diabetes mortality. 275,904 participants were included in this study. During a median follow-up of 5.56 years, 40,038 deaths were identified, including 1,179 from diabetes. Higher RC variability was associated with higher diabetes mortality. Participants in the highest quartile (Q4) of RC variability had substantially increased risk of diabetes mortality compared with those in the lowest quartile (Q1). Adjusted hazard ratios (HRs) (95% CI) for Q4 vs Q1 were 1.42 (1.21–1.67) for SD, 1.21 (1.03–1.43) for CV, 1.44 (1.22–1.71) for ARV, and 1.24 (1.05–1.47) for VIM. In sensitivity analyses restricted to participants without diabetes at baseline, the associations were no longer significant, but they were considerably more pronounced among those with diabetes. Compared with participants who maintained stable RC levels, both increased RC (HR 1.26; 95% CI 1.09–1.45) and decreased RC (HR 1.24; 95% CI 1.08–1.42) were associated with higher diabetes mortality risk. Greater RC variability, including both increased and decreased relative to stable levels, was associated with a higher risk of diabetes mortality. This association could be driven by those with pre-existing diabetes. Repeated measurements of RC may offer prognostic information for risk stratification among older adults.