Risk Factors Associated with Corneal Nerve Fiber Length Reduction in Patients with Type 2 Diabetes

Background: Diabetic neuropathy affects almost half of diabetic patients, yet the relative contributions of metabolic, vascular and clinical factors remain controversial. We aimed to investigate which risk factors are more associated with reduced corneal nerve fiber length (CNFL). Methods: This is a cross-sectional study of 30 patients with type 2 diabetes. We assessed metabolic parameters (HbA1c, lipids), vascular measurements (Doppler ultrasonography of carotid and ophthalmic arteries, central vessel density measured by optical coherence tomography angiography), and corneal epithelial thickness. We explored the data using network analysis, then applied penalized mixed-effect regression (in which β represents the standardized coefficients with mean 0 and unit standard deviation), followed by generalized additive models and polynomial transformations. Results: Penalized regression identified vascular parameters as dominant predictors: carotid plaques (β = -0.609) and intima-media thickness (β = -0.574) showed the strongest associations with CNFL. Traditional metabolic markers including HbA1c failed to meet selection thresholds. Bifurcation velocity (β = -0.313) and corneal sensitivity measures (β = 0.278-0.135) were also significant. The non-linear modeling showed complex vascular-structural interactions. Conclusions: Vascular compromise, particularly carotid disease, had the highest association with CNFL in our cohort. Thus, our study reports a higher effect of vascular parameters than HbA1c in patients with a longer history of diabetes. This may reflect the progression of diabetic complications, where initial metabolic insults are followed by vascular pathology as the primary driver of end-organ damage. Our findings highlight the need for carotid artery screening in diabetic patients for a better estimation of the neuropathy risk.