HDL cholesterol and clinical outcomes in diabetes mellitus

Abstract

Abstract Aims HDL cholesterol (HDL-C) has been thought to protect against cardiovascular disease (CVD), whereas a U-shaped association of both low and extremely high HDL-C with a high mortality risk has been increasingly reported in recent years. However, whether this U-shaped association is universal regardless of the individual’s clinical background, including lifestyle diseases, remains unclear. We examined whether fasting plasma glucose modifies the U-shaped association between the HDL-C level and clinical outcomes. Methods and results This retrospective observational cohort study analysed data from the JMDC Claims Database between 2005 and 2021 for 3 282 389 participants without a history of CVD. The median age was 44 years (IQR, 36–51), and 1 878 164 participants (57.2%) were men. The median HDL-C level was 62 (IQR 52–74) mg/dL. The study participants were categorized according to fasting plasma glucose (FPG) levels (<100 mg/dL, 100–125 mg/dL, and ≥126 mg/dL). The primary endpoint was composite CVD outcome, consisting of myocardial infarction, stroke, and all-cause death. During a mean follow-up period of 1181 ± 932 days, 35 233 composite CVD events were recorded. The association between low HDL-C and CVD risk increased with the FPG level, and the relationship of high HDL-C with CV outcome was prominent only in people with diabetes mellitus. A similar relationship was observed in the individual subgroups and in each CV outcome. Conclusion The U-shaped association between HDL-C and clinical outcomes was amplified with worsening glucose tolerance, suggesting a potential interaction between HDL-C levels and glycaemic status on clinical outcomes.