Home    中文  
 
  • Search
  • lucene Search
  • Citation
  • Fig/Tab
  • Adv Search
Just Accepted  |  Current Issue  |  Archive  |  Featured Articles  |  Most Read  |  Most Download  |  Most Cited

Chinese Journal of Ophthalmologic Medicine(Electronic Edition) ›› 2025, Vol. 15 ›› Issue (03): 141-148. doi: 10.3877/cma.j.issn.2095-2007.2025.03.003

• Original Article • Previous Articles     Next Articles

Analysis of retinal vascular fractal dimension in glaucoma and stroke

Li Zhang1,(), Shuang Yu2, Ningli Wang1   

  1. 1Beijing Institute of Ophthalmology, Beijing Ophthalmology & Visual Sciences Key Laboratory, Beijing Tongren Hospital, Capital Medical University, Beijing 100005, China
    2Tencent HealthCare, Tencent Jarvis Laboratory, Shenzhen 515899, China
  • Received:2025-03-19 Online:2025-06-28 Published:2025-09-02
  • Contact: Li Zhang

Abstract:

Objective

The aim of this study is to explore the changes in retinal vascular fractal dimension in patients with glaucoma and stroke.

Methods

From April 2020 to October 2020, 95 patients (95 eyes) with glaucoma who were treated at Beijing Tongren Hospital affiliated to Capital Medical University, 97 patients (97 eyes) with stroke who were diagnosed in the Neurology Department of Beijing Tongren Hospital, 100 healthy subjects (100 eyes) were recruited in the same period. Among them, there were 51 male patients (51 eyes) and 44 female patients (44 eyes) with an average age of (55.6±7.8) years (ranging from 46 to 69 years) in the glaucoma group; 63 male patients (63 eyes) and 34 female patients (34 eyes) with an average age of (64.7±5.6) years (ranging 58 to 73 years) in the stroke group; 50 male cases (50 eyes) and 50 female cases (50 eyes) with an average age of (52.4±6.4) years (ranging from 40 to 60 years) in the control group. All subjects underwent routine ophthalmic examinations, fundus photography, visual field and optical coherence tomography examinations, and computer-aided image analysis and measurements of retinal vessel diameter width, vessel curvature, and vessel fractal dimension were applied. The measurement values of retinal nerve fiber layer (RNFL) thickness and retinal vascular parameters were tested using Kolmogorov Smirnov method to conform normal distribution, expressed as ±s, and compared between multiple groups using one-way analysis of variance. When the differences were statistically significant, LSD was further used for comparison between two groups. The relationship between retinal vascular parameters and RNFL thickness was analyzed using Pearson correlation analysis.

Results

The average RNFL thickness in the glaucoma group, stroke group, and control group was (75.3±12.6) μm, (89.5±8.7) μm, and (103.6±6.2) μm, respectively, with statistically significant differences (F=148.516, P<0.05). The average RNFL thickness around the disc decreased in glaucoma group and stroke group, there was statistically significant difference between the glaucoma group and the stroke group compared to the control group (t=-4.362, -3.216; P<0.05). The arterial to venous width ratio in Zone B+ C, arterial to venous width ratio in Zone B, 6 larger arterial widths located in Zone B+ C, 6 larger arterial widths located in Zone B, 6 larger venous widths located in Zone B+ C, 6 larger venous widths located in Zone B, all vessel simple bending, all vessel curvature bending, arterial simple bending, arterial curvature bending, venous simple bending, venous curvature bending, all vessel fractal dimension, arterial fractal dimension, and venous fractal dimension were 0.552±0.216, 0.530±0.137, (8.491±2.039) pixels, (8.804±2.136) pixels, respectively, for the control group, glaucoma group, and stroke group (15.959±2.138) pixels, (15.934±2.962) pixels, 1.057±0.984, 0.033±0.002, 1.083±0.169, 0.027±0.026, 1.069±0.187, 0.042±0.006, 1.527±0.148, 1.295±0.065, 1.344±0.058, 0.558±0.111, 0.567±0.104, (8.852±1.931) pixels, (8.923±1.872) pixels, (16.015±2.787) pixels, (15.891±2.921) pixels, 1.069±0.251, 0.003±0.001 1.044±0.021, 0.003±0.001, 1.068±0.041, 0.044±0.002, 1.529±0.032, 1.286±0.051, 1.342±0.049, 0.609±0.108, 0.774±0.143, (10.082±2.151) pixels, (10.417±2.223) pixels, (16.548±2.885) pixels, (16.739±2.768) pixels, 1.221±1.378, 0.331±0.024, 1.262±1.605, 0.029±0.015, 1.082±0.209, 1.932±0.137 1.544±0.156, 1.331±0.039, and 1.367±0.035, showed statistically significant differences (F=108.973, 117.136, 104.762, 119.531, 113.538, 121.539, 119.968, 134.472, 125.162, 104.167, 76.106, 136.262, 109.421, 116.632, 43.279; P<0.05). Compared between the glaucoma group and the control group, vascular parameters of retinal arteriovenous width ratio (Zone B+ C), arterial width in Zone B+ C, arterial width in Zone B, retinal vein width in Zone B+ C, retinal vein width in Zone B, simple bending of all retinal blood vessels, curvature bending of all retinal blood vessels, simple bending of retinal arteries, curvature bending of retinal veins, fractal dimension of retinal blood vessels, and fractal dimension of retinal arteries, had shown that the differences were statistically significant (t=-3.326, -3.761, -3.964, -2.237, -2.453, -2.447, -3.895, -2.963, -3.871, -2.037, -2.363; P<0.05). The fractal dimension of retinal blood vessels in glaucoma patients was positively correlated with the thickness of RNFL around the disc, and the correlation was statistically significant (r=0.783, P<0.05). Compared between the stroke group and the control group, vascular parameters of retinal arteriovenous width ratio (Zone B+ C), arterial width in Zone B+ C, arterial width in Zone B, retinal vein width in Zone B+ C, retinal vein width in Zone B, simple bending of all retinal blood vessels, curvature bending of all retinal blood vessels, simple bending of retinal arteries, curvature bending of retinal veins, fractal dimension of retinal blood vessels, and fractal dimension of retinal arteries, had shown that the differences were statistically significant (t=-2.741, -3.153, -2.930, -2.041, -2.726, -2.323, -3.881, -3.267, -2.965, -2.024, -2.473; P<0.05). The fractal dimension of retinal blood vessels in stroke patients was positively correlated with the thickness of RNFL around the disc, and the correlation is statistically significant (r=0.645, P<0.05). The lower the fractal dimensionality of retinal vascular analysis in stroke patients, the lower the average RNFL thickness.

Conclusions

Patients with glaucoma and patients with stroke have narrower retinal vessel diameter, decreased retianl vascular bending, and decreased retinal vascular fractal dimension. The parameters of retinal vascular in stroke patients show a trend similar to glaucoma patients. The fractal dimension of retinal vascular in glaucoma is correlated with the severity of glaucoma, lower retinal vascular fractal dimension and RNFL thickness are associated with stroke.

Key words: Fractal dimension, Retinal vascular, Retinal nerve fiber layer, Glaucoma, Stroke

Copyright © Chinese Journal of Ophthalmologic Medicine(Electronic Edition), All Rights Reserved.
Tel: 0086-10-58269646 E-mail: zhykyxzz@163.com
Powered by Beijing Magtech Co. Ltd