International Journal of Retina and Vitreous最新文献

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.

Background: Swept-source OCT (SS-OCT) and OCT angiography (SS-OCTA) enable high-resolution assessment of retinal and choroidal biomarkers in diabetic macular edema (DME). However, prospective analyses of how these biomarkers correlate before and after anti-VEGF therapy in treatment-naïve eyes are limited. The aim of this study was to prospectively evaluate biomarker correlations following aflibercept treatment using state-of-the-art, high-resolution imaging with SS-OCT and SS-OCTA during 4 months of follow-up.

Methods: This was a prospective interventional case series that included 28 eyes from 25 treatment-naïve patients with DME. All eyes received three monthly intravitreal aflibercept injections. Patients were reassessed one month after the loading phase (4-month visit). The evaluated biomarkers included best-corrected visual acuity (BCVA), central macular thickness (CMT), central choroidal thickness (CCT), vessel density (VD), and avascular area of the superficial plexus (AASP) and deep plexus (AADP). Pre- and posttreatment values were compared, and correlations were analyzed using Pearson's or Spearman's methods.

Results: Significant changes in BCVA (0.7250 ± 0.23 to 0.3957 ± 0.21; p < 0.000), CMT μm (339.04 ± 66.19 to 265.21 ± 55.75; p < 0.000), CCT μm (221.71 ± 69.69 to 209.07 ± 70.92; p < 0.000), VD (17.90 ± 7.82 to 15.35 ± 5.80; p < 0.038), AASP μm² (235,374 ± 91,299 to 157,326 ± 77,815; p < 0.000) and AADP μm² (1996,335 ± 1,000,047 to 362,161 ± 277,225; p < 0.000) were detected. Very high correlations were observed for the CCT pre vs CCT post (r = 0.98; p < 0.001), and AADP pre vs AADP reduction (r = -0.93; p < 0.001), high correlation: VD pre vs VD reduction (r = -0.72; p < 0.001. In total, 2 correlations were very high, 1 high, 4 moderate, and 9 were low; all the correlations were statistically significant.

Conclusions: Treatment-naïve DME eyes treated with aflibercept showed significant structural, vascular, and functional improvements, with several baseline biomarkers acting as potential predictive indicators of posttreatment outcomes. The very high correlation of CCT and AADP and the high correlation of VD suggest that SS-OCT and SS-OCTA can provide clinically useful information for identifying how patients will respond to treatment. These correlations support the role of SS-OCT and SS-OCTA not only as diagnostic tools but also as potential predictive indicators of therapeutic response, facilitating more personalized DME management in clinical practice. Notably, this is the first prospective study to evaluate correlations between SS-OCT and SS-OCTA biomarkers in naïve DME eyes treated with aflibercept.

Background: In the early1990s, neovascular age-related macular degeneration (nAMD) was the leading cause of irreversible vision loss in older adults, yet its molecular basis remained unknown. In 1992, a hypothesis was proposed in which localized hypoxia could trigger vascular endothelial growth factor (VEGF)-mediated choroidal neovascularization in nAMD. Although hypoxia was recognized in ischemic retinopathies, nAMD was not considered a hypoxia-mediated retinal vascular disease.

Main body: In 1994, this hypothesis was tested using antigen retrieval immunohistochemistry on paraffin-embedded whole human eye sections with early nAMD. The study demonstrated strong VEGF immunoreactivity in the retinal pigment epithelium in the macular area but not in normal control eyes, providing the first direct histopathologic evidence of VEGF expression at the site of disease in intact human eyes with early nAMD. Until that point, the role of VEGF in ischemic retinopathies was being uncovered, but its involvement in early-stage nAMD had not yet been demonstrated.

Conclusion: The 1992-1994 work established both the hypothesis and the first direct tissue evidence linking VEGF to early nAMD. This discovery, made just over a decade before the advent of anti-VEGF therapy, anticipated one of ophthalmology's most transformative achievements, preserving vision for millions worldwide.