Risk prediction model for long-term visual acuity recovery in macular edema secondary to branch retinal vein occlusion patients after combination therapy
10.3760/cma.j.cn115455-20240708-00580
- VernacularTitle:视网膜分支静脉阻塞继发黄斑水肿患者联合用药后远期视力恢复的风险预测模型
- Author:
Ruifen ZHANG
1
;
Fang YUAN
Author Information
1. 西安爱尔古城眼科医院眼科，西安 710000
- Publication Type:Journal Article
- Keywords:
Macular edema;
Branch retinal vein occlusion;
Long-term vision;
Risk prediction model
- From:
Chinese Journal of Postgraduates of Medicine
2025;48(7):628-634
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To analyze the possible factors affecting the long-term visual acuity of macular edema secondary to branch retinal vein occlusion (BRVO-ME) after combination therapy, and construct a risk prediction model.Methods:The data of 125 patients with BRVO-ME who were treated in Xi′an Aier Ancient City Eye Hospital from December 2020 to December 2023 and completed 6-month follow-up were retrospectively analyzed. All patients were treated with ranibizumab combined with dexamethasone intravitreal implants until the central macular thickness returned to normal. The treatment was completed and followed up for 6 months. According to the 6-month follow-up of long-term visual acuity recovery, they were divided into poor recovery group (38 cases) and good recovery group (87 cases). The general data of the two groups were collected and compared. The possible influencing factors (continuous variables) were included in the drawing of the receiver operating characteristic (ROC) curve. According to the optimal threshold, the continuous variables were divided into two categorical variables, and the possible influencing factors were further included in the Logistic regression to obtain the possible influencing factors. Based on the results of the influencing factor analysis and the ROC curve, the decision curve was drawn to verify the clinical benefit of the overall model. Based on the above-mentioned nomogram model, the prediction model was visualized.Results:The course of disease in the poor recovery group was longer than that in the good recovery group: (82.47 ± 15.51) d vs. (71.23 ± 12.45) d, the proportion of ellipsoid/external membrane interruption was higher than that in the good recovery group: 44.74% (17/38) vs. 25.29% (22/87), the initial visual acuity was worse than that in the good recovery group: (0.82 ± 0.21) logMAR vs. (0.66 ± 0.17) logMAR, and the ratio of retinal deep and shallow blood flow density was lower than that in the good recovery group: 1.02 ± 0.35 vs. 1.36 ± 0.41, the difference was statistically significant ( P<0.05). ROC curve analysis showed that the area under the curve of the course of disease, the initial visual acuity, and the ratio of retinal deep and shallow blood flow density to predict the poor recovery of long-term visual acuity in BRVO-ME patients after combined medication were 0.71, 0.73 and 0.75, respectively. The best cut-off values were 85 d, 0.72 logMAR and 1.10, respectively. Logistic regression analysis showed that the course of disease (≥ 85 d), ellipsoid/outer membrane interruption (yes), initial visual acuity (≥ 0.72 logMAR) and ratio of retinal deep and shallow blood flow density (≤1.10) were the influencing factors of poor long-term visual acuity recovery in BRVO-ME patients after combination therapy ( P<0.05). The risk prediction model was constructed by using the course of disease, ellipsoid/outer membrane interruption, initial visual acuity and ratio of retinal deep and shallow blood flow density as predictors, and the decision curve of the model was drawn. It was found that the risk threshold was in the range of 0 to 0.90 and 0.95 to 1.00, and the clinical net benefit rate provided by the model was always greater than 0. The nomogram model was constructed to visualize the risk prediction model. The results showed that the C-index of the model to predict the long-term visual acuity recovery of BRVO-ME patients after combination therapy was 0.90 (95% CI 0.85 to 0.96), which had high predictive value. Conclusions:The course of the disease, interruption of the ellipsoid/outer membrane, initial visual acuity and the ratio of retinal deep and shallow blood flow density are all influencing factors for poor long-term visual recovery in BRVO-ME patients after combination therapy. The risk prediction model constructed based on this has good predictive value for the risk of poor long-term visual recovery in patients.
