The linear association between high-density lipoprotein cholesterol and diabetic retinopathy : a cross-sectional study in diabetic patients.

Background: Diabetic retinopathy (DR) is a major microvascular complication of diabetes mellitus and a leading cause of visual impairment. While dyslipidemia has been implicated in DR pathogenesis, the relationship between high-density lipoprotein cholesterol (HDL-C) and DR remains controversial. This study aimed to investigate the association between HDL-C levels and DR prevalence in diabetic patients type 2 diabetes mellitus (T2DM). Methods: This study is the second analysis based on a cross-sectional studv. A total of 2001 (858 men and 1143 women) diabetic patients who visited the diabetic clinic in the Internal Medicine out-patient departments of two hospitals in southern Taiwan between April 2002 and November 2004. Demographic and clinical data were collected, and serum HDL-C levels were measured. The association between HDL-C and DR was analyzed using multivariate logistic regression, accounting for potential confounders. Additionally, we explored the potential correlation between HDL-C and DR through a smooth curve fitting approach, utilizing a generalized additive model (GAM). and a generalized additive model (GAM). Results: Among the 2001 participants, 701 (35.0%) were diagnosed with DR. Our findings demonstrated a significant inverse linear relationship between HDL-C levels and the presence of DR. Higher HDL-C was inversely associated with diabetic retinopathy (DR). In continuous analyses, each 10 mg/dL increase in HDL-C corresponded to lower odds of DR across all models, including Model 3 (OR 0.92; 95% CI 0.84-0.99; P = 0.027).Additionally, analysis of HDL-C levels by tertiles revealed that participants in the highest tertile (53.0-99.0 mg/dL) had a lower prevalence of DR, with an OR of 0.78 (95% CI: 0.62-0.97; P = 0.029) in Model 1; this association was borderline in Model 2 (OR: 0.80; 95% CI: 0.63-1.01; P = 0.055) and non-significant in Model 3 (OR: 0.86; 95% CI: 0.66-1.09; P = 0.209). In categorical analyses (reference: ≤40 mg/dL), participants with HDL-C ≥ 60 mg/dL exhibited a significantly lower prevalence of DR in Model 1 (OR: 0.73; 95% CI: 0.56-0.97; P = 0.028), a borderline association in Model 2 (OR: 0.76; 95% CI: 0.57-1.01; P = 0.059), and a non-significant association in Model 3 (OR: 0.83; 95% CI: 0.61-1.14; P = 0.259). Conclusions: This study provides evidence of a linear association that elevated HDL-C levels are associated with decreased odds of DR in diabetic patients. Future research should further focus on elucidating the mechanisms underlying this association and its implications for therapeutic strategies. Supplementary Information: The online version contains supplementary material available at 10.1007/s40200-025-01728-z.