Lipid biomarkers and diabetic retinopathy: a propensity score matching lipidomic study

Lipid metabolites may play an important role in the pathogenesis of diabetic retinopathy (DR); however, their utility as biomarkers has not been comprehensively investigated. This study aims to characterize the relationship between plasma lipidomic profiles and DR and to evaluate the potential utility of lipid metabolites for DR detection. This propensity score matching study included 69 matched pairs of individuals with type 2 diabetic mellitus (T2DM) with and without DR. Plasma lipid metabolites were assessed using an ultra-high-performance liquid chromatography-high resolution tandem mass spectrometry platform. Principal component analysis, orthogonal partial least squares discriminant analysis, and least absolute shrinkage and selection operator regression were integrated to identify lipid biomarkers associated with DR. A lipid risk score (LRS) was constructed based on the identified lipid biomarkers. Restricted cubic spline models and area under the receiver operating characteristic curve (AUC) analysis were applied to evaluate the association between the LRS and DR. A total of 499 and 436 lipid metabolites were identified in the positive and negative ion modes, respectively. Nine differential lipid metabolites (ACar 8:0, SM d38:0, TAG 53:4, OxPC 36:4 + 1O, LPG 18:1, FA 22:5, AcylGlcADG 64:16, PE 38:5, and LPC 17:0) were retained as potential lipid biomarkers for DR following feature selection procedures. The LRS demonstrated a positive association with DR. In the training set, the AUC was 0.94 (95% confidence interval[CI]: 0.89, 0.99), with a sensitivity of 84.78% and a specificity of 91.3%. In the validation set, the AUC was 0.81 (95% CI: 0.68, 0.94), with a sensitivity of 78.26% and a specificity of 82.61%. Nine potential lipid biomarkers associated with DR were identified among individuals with T2DM. These findings provide a basis for further investigation into lipid metabolite alterations associated with DR and their potential application in the early identification of individuals at risk for DR.