Association of Albumin-To-Creatinine Ratio With Diabetic Retinopathy Among US Adults (NHANES 2009-2016).

Abstract

Objective: This study investigates the relationship between the albumin-to-creatinine ratio and diabetic retinopathy (DR) in US adults using NHANES data from 2009 to 2016. This study assesses the predictive efficacy of the urinary serum albumin-to-creatinine ratio (UACR/SACR Ratio) against traditional biomarkers such as the serum albumin-to-creatinine ratio (SACR) and urinary albumin-to-creatinine ratio (UACR) for evaluating DR risk. Additionally, the study explores the potential of these biomarkers, both individually and in combination with HbA1c, for early detection and risk stratification of DR.

Methods: This cross-sectional study analysed data from 2594 diabetic adults in the National Health and Nutrition Examination Survey (NHANES 2009-2016). Multivariate logistic regression models, adjusted for demographic (sex, age, race and education) and clinical factors (WBC, PLT, RDW, HbA1c, HBP and CHD), examined the associations between biomarkers and DR. Biomarkers were analysed both continuously and in quartiles to assess dose-response relationships. Receiver operating characteristic (ROC) curve analysis evaluated the predictive accuracy of individual biomarkers and combined models.

Results: Elevated SACR levels were inversely related to DR risk, while UACR showed a positive correlation. The UACR/SACR ratio demonstrated superior predictive capability for DR compared to SACR and UACR alone. The most accurate predictive model combined SACR, UACR, UACR/SACR ratio and HbA1c (AUC = 0.614), highlighting DR development complexity. Subgroup analyses revealed stronger associations in participants aged < 60 years and those with hypertension (both p < 0.05), with more pronounced effects observed in males and Mexican Americans, while lifestyle factors showed no significant modifying effect.

Conclusion: The UACR/SACR Ratio emerged as a potentially superior DR predictor, particularly in younger patients and those with hypertension, suggesting its utility in enhancing early detection and risk stratification. Comprehensive evaluation of renal function and glycaemic control biomarkers, considering age- and comorbidity-specific patterns, could improve DR risk prediction and management. Future longitudinal studies should validate these findings, particularly in identified high-risk subgroups, and investigate underlying mechanisms, potentially advancing personalised DR prediction and management strategies.