Prediction of Diabetic Retinopathy Progression Based on Logistic Regression Model

Diabetic retinopathy (DR) remains a leading cause of blindness in the working-age diabetic population. DR progression often occurs asymptomatically, and current screening methods do not always detect the onset of retinal deterioration in a timely manner. This creates a strong need for risk prediction models that can identify high-risk patients in advance and enable prompt preventive treatment. Moreover, the pathogenic mechanisms of DR may differ between type 1 and type 2 diabetes (neurodegeneration and inflammation vs. microangiopathy), which justifies developing separate progression models for each diabetes type.

Purpose: tо develop and validate multivariable logistic regression models predicting 1-year progression of diabetic retinopathy (DR) in type 1 diabetes (T1D), type 2 diabetes (T2D), and a combined cohort, and to compare their accuracy and key predictors.

Patients and methods. In this prospective study, 731 patients (731 eyes) without proliferative DR underwent optical coherence tomography (OCT), OCT angiography (OCTA), and systemic evaluation. A total of 102 candidate predictors were considered. Separate multivariable logistic models were built for T1D, T2D, and the combined cohort (75 % training / 25 % validation; class balancing). Performance was assessed by macro F-measure, sensitivity, specificity, and negative predictive value (NPV).

Results. The T1D model achieved a macro F-score of 0.96 with 100 % sensitivity and 97.3 % specificity (NPV 100 %). The T2D model yielded a macro F-score of 0.85 with 90.5 % sensitivity and 87.8 % specificity (NPV 97 %). The combined model showed a macro F-score of 0.91 with 91.2 % sensitivity and 94.6 % specificity (NPV 97.9 %). Key predictors were: for T1D— hyperreflective foci (HRF), relative ganglion cell complex thickness (GCC %), and central retinal thickness; for T2D—fractal dimension in the superficial capillary plexus (FD-SCP), GCC and GCL+ thicknesses, and estimated glomerular filtration rate (eGFR); for the combined model—HRF, GCC, GCL+, and eGFR.

Conclusions. Logistic regression models were developed for T1D, T2D, and a combined cohort; validation confirmed high predictive accuracy and delineated distinct key predictors—predominantly neurodegenerative and neuroinflammatory biomarkers in T1D, and neurodegenerative, vascular, and systemic indicators in T2D.

1. Kropp M, Golubnitschaja O, Mazurakova A, Koklesova L, Sargheini N, Vo TKS, de Clerck E, Polivka J Jr, Potuznik P, Polivka J, Stetkarova I, Kubatka P, Thumann G. Diabetic retinopathy as the leading cause of blindness and early predictor of cascading complications-risks and mitigation. EPMA J. 2023 Feb 13;14(1):21–42. doi: 10.1007/s13167-023-00314-8.

2. Teo ZL, Tham YC, Yu M, Chee ML, Rim TH, Cheung N, Bikbov MM, Wang YX, Tang Y, Lu Y, Wong IY, Ting DSW, Tan GSW, Jonas JB, Sabanayagam C, Wong TY, Cheng CY. Global Prevalence of Diabetic Retinopathy and Projection of Burden through 2045: Systematic Review and Meta-analysis. Ophthalmology. 2021 Nov;128(11):1580–1591. doi: 10.1016/j.ophtha.2021.04.027.

3. Neroev VV, Zaitseva OV, Mikhailova LA. Diabetic retinopathy prevalence in the Russian Federation according to all-Russia statistics. Russian Ophthalmological Journal. 2023;16(3):7–11 (In Russ.). doi: 10.21516/2072-0076-2023-16-3-7-11.

4. Antonetti DA, Klein R, Gardner TW. Diabetic retinopathy. N Engl J Med. 2012 Mar 29;366(13):1227–1239. doi: 10.1056/NEJMra1005073.

5. Wilkinson CP, Ferris FL 3rd, Klein RE, Lee PP, Agardh CD, Davis M, Dills D, Kampik A, Pararajasegaram R, Verdaguer JT; Global Diabetic Retinopathy Project Group. Proposed international clinical diabetic retinopathy and diabetic macular edema disease severity scales. Ophthalmology. 2003 Sep;110(9):1677–1682. doi: 10.1016/S0161-6420(03)00475-5.

6. Fayzrakhmanov RR, Shishkin MM, Saraeva SN. A modern perspective on the pathogenesis of proliferative diabetic retinopathy. Russian Annals of Ophthalmology. 2025;141(1):92–98 (In Russ.). doi: 10.17116/oftalma202514101192.

7. Midena E, Torresin T, Velotta E, Pilotto E, Parrozzani R, Frizziero L. OCT Hyperreflective Retinal Foci in Diabetic Retinopathy: A Semi-Automatic Detection Comparative Study. Front Immunol. 2021 Apr 22;12:613051. doi: 10.3389/fimmu.2021.613051.

8. Gong X, Wang W, Li W, Jin L, Wang L, Meng J, Xiong K, Li Y, Han X, Liang X, Huang W. Association between renal function and retinal neurodegeneration in Chinese patients with type 2 diabetes mellitus. Ann Transl Med. 2021 Apr;9(7):560. doi: 10.21037/atm-20-6957.

9. Gundogan FC, Akay F, Uzun S, Yolcu U, Çağıltay E, Toyran S. Early Neurodegeneration of the Inner Retinal Layers in Type 1 Diabetes Mellitus. Ophthalmologica. 2016;235(3):125–132. doi: 10.1159/000442826.

10. Shah J, Tan B, Wong D, Abdul Gani NFB, Hu Q, Liu X, Chua J. Evaluation of thickness of individual macular retinal layers in diabetic eyes from optical coherence tomography. Sci Rep. 2024 Aug 2;14(1):17909. doi: 10.1038/s41598-024-68552-y.

11. Hashmi S, Lopez J, Chiu B, Sarrafpour S, Gupta A, Young J. Fractal Dimension Analysis of OCTA Images of Diabetic Retinopathy Using Circular Mass-Radius Method. Ophthalmic Surg Lasers Imaging Retina. 2021 Mar;52(3):116–122. doi: 10.3928/23258160-20210302-01.

12. Sun Z, Tang F, Wong R, Lok J, Szeto SKH, Chan JCK, Chan CKM, Tham CC, Ng DS, Cheung CY. OCT Angiography Metrics Predict Progression of Diabetic Retinopathy and Development of Diabetic Macular Edema: A Prospective Study. Ophthalmology. 2019 Dec;126(12):1675–1684. doi: 10.1016/j.ophtha.2019.06.016. Epub 2019 Jun 26. Erratum in: Ophthalmology. 2020 Dec;127(12):1777. doi: 10.1016/j.ophtha.2020.09.022.

13. Park HC, Lee YK, Cho A, Han CH, Noh JW, Shin YJ, Bae SH, Kim H. Diabetic retinopathy is a prognostic factor for progression of chronic kidney disease in the patients with type 2 diabetes mellitus. PLoS One. 2019 Jul 29;14(7):e0220506. doi: 10.1371/journal.pone.0220506.

14. Fursova AZh, Derbeneva AS, Vasilyeva MA, Tarasov MS, Nikulich IF, Gamza YuA. Structural and microvascular changes in the retina and choroid in patients with chronic kidney disease. Russian Annals of Ophthalmology. 2021;137(6):99–108 (In Russ.). doi: 10.17116/oftalma202113706199.

15. Cho A, Park HC, Lee YK, Shin YJ, Bae SH, Kim H. Progression of Diabetic Retinopathy and Declining Renal Function in Patients with Type 2 Diabetes. J Diabetes Res. 2020 Jul 26;2020:8784139. doi: 10.1155/2020/8784139.

16. Bianco L, Arrigo A, Aragona E, Antropoli A, Berni A, Saladino A, Battaglia Parodi M, Bandello F. Neuroinflammation and neurodegeneration in diabetic retinopathy. Front Aging Neurosci. 2022 Aug 16;14:937999. doi: 10.3389/fnagi.2022.937999.