International Journal of Retina and Vitreous最新文献

Background: To identify baseline predictors of anatomical response after intravitreal bevacizumab in polypoidal choroidal vasculopathy (PCV) and to evaluate anatomical and visual outcomes over 12 months.

Methods: This retrospective study included 81 eyes with PCV treated with three consecutive monthly intravitreal bevacizumab injections. Patients were classified as good or poor responders based on optical coherence tomography (OCT) findings at month 3. Baseline demographic, clinical, and imaging parameters were compared between groups, and predictors of poor response were identified using multivariable regression. Visual acuity (VA), central retinal thickness (CRT), subretinal fluid (SRF), and pigment epithelial detachment (PED) height were analyzed within and between groups through month 12.

Results: Forty eyes (49.4%) were poor responders and 41 (50.6%) were good responders. Greater baseline SRF height and thinner SFCT were independent predictors of poor response (RR per 100 μm: 1.27 and 0.71 respectively; AUC = 0.74). Good responders showed significant VA improvement (0.64 ± 0.08 vs. 0.48 ± 0.06 logMAR, p < 0.001) and CRT reduction (419.5 ± 24.3 vs. 285.5 ± 14.1 μm, p < 0.001) at month 3, and these were maintained at month 12. Among good responders, 75% continued bevacizumab injections (non-switch group) and sustained favorable outcomes. In contrast, 86% of poor responders switched to other medications (switch group), achieving significant VA gain, CRT reduction, and further SRF with PED reduction by month 12. The non-switch group required fewer injections (5.8 ± 2.5 vs. 7.8 ± 3.0 injections; p < 0.001) and achieved longer treatment intervals (11.7 ± 9.0 vs. 7.8 ± 3.0 weeks; p = 0.046) than the switch group.

Conclusions: Approximately half of PCV eyes demonstrated a favorable response to bevacizumab monotherapy. Baseline SRF height and SFCT were significant predictors of treatment response. Early assessment after three monthly bevacizumab injections allowed for the differentiation between relatively mild and a more severe disease requiring more frequent and sustained anti-VEGF injections. Switching of anti-VEGFs was associated with improved outcomes in poor responders.

Purpose: To determine the associations between axial length (AL) and optical coherence tomography angiography (OCTA)-based eye-related parameters.

Methods: A prospective cross-sectional study of the right eye of 6953 individuals between the ages 50 and 97, with a range of AL from 19.1 to 31.7 mm was performed. Central subfield thickness (CST), foveal avascular zone (FAZ) and mean vessel density (mVD) were measured by OCTA. Associations between OCTA-based parameters, eye-related and systemic-related parameters were assessed using one-way ANCOVA, multivariable linear regression analysis and multiple regression analysis.

Results: AL and OCTA-based parameters were negatively correlated with age. Increased AL was associated with increased CST (ß=3.59), decreased FAZ (ß=0.02), and decreased mVD (ß=-0.4) (all p = 0.000). Better best corrected visual acuity (BCVA) was associated with larger FAZ (ß=-0.07) and higher mVD (ß=-3.160) (p = 0.002, p = 0.000). Hypertension and diabetes mellitus correlated with increased mVD (ß=0.5) and decreased mVD (ß=-0.18) respectively (p = 0.000, p = 0.023).

Conclusion: Increased AL is associated with lower mVD which may be a potential factor between myopia and poor BCVA.

Background: Cystoid macular edema (CME) is a leading cause of vision loss in patients with retinitis pigmentosa (RP). The present study was aimed to CME in patients with RP using deep learning (DL) models based on the analysis of the optical coherence tomography (OCT) images.

Methods: In this cross-sectional study, a total of 1,318 OCT scans of 296 eyes from RP patients were analyzed with scans grouped based on the presence (670 images) or absence (648 images) of CME. We used Spectral-Domain OCT (SD-OCT) to measure central foveal thickness and detect retinal abnormalities, including subclinical CME in the study groups. The dataset was stratified and divided into training and testing sets using a subject-wise split at an 80:20 ratio using the scikit-learn library. Resnet-34 and ResNet-18 model architectures were developed to automatically detect CME in RP patients, and their performance was evaluated and compared with other DL algorithms.

Results: Fine-tuning pretrained ResNet-34 and ResNet-18 models achieved an accuracy of 99.25%, 98.75%, F1-score of 99.26%, 98.77% and ROC of 99% and non-pretrained ResNet-34 and ResNet-18 achieved an accuracy of 80.64%, 82.13%, F1-score of 83.88%, 84.87% and ROC of 80%, 82% in detection of CME in RP patients.

Conclusion: This study was the first to apply DL algorithms to diagnose and manage CME in RP patients using OCT images. Pretrained ResNet models, particularly ResNet-34 with 99.25% accuracy, outperformed non-pretrained ResNet-34 with 80.64% accuracy. These results underscore the potential of pretrained models to aid in detection CME and supporting remote healthcare services.