Development and external validation of a predictive model for type 2 diabetic retinopathy

Area of Science:

• Ophthalmology
• Endocrinology
• Medical Informatics

Background:

• Diabetic retinopathy (DR) is a leading cause of irreversible blindness in adults.
• Early detection and intervention are crucial for managing DR and preserving vision.
• Existing methods for DR risk prediction require enhancement for clinical utility.

Purpose of the Study:

• To develop and externally validate a predictive nomogram for early diabetic retinopathy (DR) risk in type 2 diabetes mellitus (T2DM) patients.
• To identify key clinical and biochemical predictors of DR.
• To assess the nomogram's predictive performance and clinical applicability.

Main Methods:

• Retrospective study of 2381 T2DM patients for model development and 962 T2DM patients for external validation.
• Utilized Least Absolute Shrinkage and Selection Operator (LASSO) and multivariate logistic regression to identify independent predictors.
• Nomogram performance evaluated using Receiver Operating Characteristic (ROC) curves, calibration curves, and Decision Curve Analysis (DCA).

Main Results:

• A nomogram was established using Neutrophil, 25-hydroxyvitamin D3 [25(OH)D3], Duration of T2DM, hemoglobin A1c (HbA1c), and Apolipoprotein A1 (ApoA1).
• The nomogram achieved high predictive accuracy with Areas Under the Curve (AUC) of 0.834 in the development group and 0.851 in the external validation group.
• The nomogram demonstrated excellent calibration and clinical usefulness via DCA.

Conclusions:

• The developed and validated nomogram is a reliable tool for predicting DR risk in T2DM patients.
• Early identification of high-risk individuals enables timely clinical intervention.
• This model holds significant potential for improving long-term visual health outcomes in diabetes patients.