AI-Driven Retinal Imaging for Cardiovascular Risk Stratification in Diabetes—Findings from Ret-CAC Screening at GluCare

Introduction and Objective:

 Fundus exams are routine in diabetes management, but retinal analysis for CVD risk isn’t standard. Retinal microvascular changes also provide valuable insights into systemic vascular health, presenting an opportunity to enhance CVD risk assessment. AI-driven retinal imaging, such as Ret-CAC, offers a non-invasive method to detect coronary artery calcification and assess CVD risk. This study explores clinical and metabolic factors linked to high CVD risk in T2D patients and evaluates Ret-CAC as a screening tool for early intervention.

Methods: 

This retrospective observational study included patients with T2D aged ≥40 years who underwent Ret-CAC screening at GluCare (n=115). Exclusion criteria included prior CVD history or incomplete imaging data. Independent variables included age, sex, BMI, HbA1c, lipid profile, blood pressure, and lifestyle factors. The primary outcome was CVDbrisk classification (low, moderate, high) based on Ret-CAC results.

Results: 

The mean age of participants was 49.79 ± 9.44 years, with 74.78% males. Hyperlipidemia was diagnosed in 96.43%, with 83.48% receiving statin treatment. Around 25.2% were classified as high-risk, 36.5% as moderate and 38.3% as low risk groups for CVD disease. Unadjusted analysis showed that higher age, longer diabetes duration, presence of hypertension, and reduced eGFR were significant predictors of high CVD risk. After adjustment, age remained a significant predictor for moderate (OR: 1.163; 95% CI: 1.064, 1.271) and high risk (OR: 1.356; 95% CI: 1.201, 1.532) when compared with low risk. Females were associated with lower odds of high risk (OR: 0.039; 95% CI: 0.004, 0.366). Lower eGFR persisted as a predictor for high CVD risk (OR: 0.957; 95% CI: 0.926, 0.990).

Conclusion: 

Ret-CAC identified significant CVD risk in patients with T2D, with age and eGFR as key predictors. AI-driven retinal imaging may enhance early risk stratification and guide preventive interventions.