Smaller HDL Particles and ABCA1 Cholesterol Efflux Capacity Predict Incident Proliferative Retinopathy Only among Men with Type 1 Diabetes

Clinical studies have provided evidence for a potential role of dyslipidemia in the development of retinopathy in type 1 diabetes. Existing evidence, however, is generally based on conventional lipid analysis and/or cross-sectional study designs. We therefore assessed the prospective association of HDL particle concentration (HDL-P) and cholesterol efflux capacity (CEC) with incident proliferative diabetic retinopathy (PDR) in a cohort of childhood-onset type 1 diabetes. HDL-P (calibrated differential ion mobility analysis) and total and ABCA1-specific CEC (validated cell-based assays) were quantified in 186 men and 185 women with type 1 diabetes free of PDR (stereo fundus photography) at baseline (mean age 25.6 years; diabetes duration, 17.1 years). Women had higher large HDL-P than men; the concentrations of extra-small, small and medium HDL-P as well as both total and ABCA1 CEC were similar by sex. During a median follow-up of 28 years, 108 men and 100 women developed PDR. In multivariable Cox models stratified by sex, HDL-C did not predict PDR incidence in either sex. However, in addition to increased HbA1c, non-HDL-C, white blood cell count and albumin excretion rate, increased extra-small (HR=0.25, 95% CI=0.09-0.67) and small (HR=0.72, 95% CI=0.55-0.94) HDL-P concentrations and ABCA1 CEC (HR=0.80, 95% CI=0.64-0.998) negatively predicted PDR in men but not women, among whom HbA1c, hypertension and white blood cell count were the only consistent significant predictors of PDR. In conclusion, while extra-small and small HDL-P concentrations and ABCA1 CEC did not differ by sex in type 1 diabetes, they only predicted incident PDR in men, a finding similar to our previously reported results for coronary artery disease. Further research is warranted to better understand the role of HDL in the pathophysiology of PDR and why this appears to differ by sex in type 1 diabetes. Disclosure T.Costacou: None. R.G.Miller: None. T.J.Orchard: None. Funding National Institutes of Health (R01HL130153, R01DK034818)