Association between the non-high-density lipoprotein cholesterol to high-density lipoprotein cholesterol ratio and cardiovascular mortality among US adults with diabetes: a prospective cohort study of NHANES 1999–2018

Background: The non-high-density lipoprotein cholesterol to high-density lipoprotein cholesterol ratio (NHHR) integrates information on atherogenic and anti-atherogenic lipoproteins. Its association with cardiovascular mortality in adults with diabetes remains unclear. Methods: We conducted a prospective cohort analysis of ten National Health and Nutrition Examination Survey (NHANES) cycles (1999-2018) linked to mortality follow-up through December 31, 2019. Diabetes was defined by self-reported diagnosis, insulin use, oral glucose-lowering medication use, or hemoglobin A1c (HbA1c) ≥6.5%. NHHR was calculated as (total cholesterol - HDL-C)/HDL-C. Survey-weighted cause-specific Cox models and restricted cubic splines were used to examine the association between NHHR and cardiovascular mortality. Results: The complete-case cohort comprised 6,581 adults with diabetes, including 622 cardiovascular deaths. Compared with Q1, the fully adjusted hazard ratios for cardiovascular mortality were 0.89 (95% confidence interval [CI], 0.67-1.19) for Q2, 1.07 (95% CI, 0.84-1.36) for Q3, and 1.31 (95% CI, 0.95-1.80) for Q4. The hazard ratio per one-quartile increase was 1.11 (95% CI, 1.00-1.22; P for trend=0.043). Restricted cubic spline analysis suggested a nonlinear association (P for overall association=0.024; P for nonlinearity=0.020), with the lowest estimated hazard at an NHHR of approximately 2.5-3.0. The direction of association was similar in sensitivity analyses, although statistical significance was attenuated. Conclusions: NHHR was nonlinearly associated with cardiovascular mortality among US adults with diabetes. Higher NHHR values above the apparent nadir were associated with a tendency toward higher cardiovascular mortality; however, the association was modest and less precise in sensitivity analyses. These findings warrant cautious interpretation and external confirmation.