Outcomes of anti-VEGF treatment for macular edema secondary to central retinal vein occlusion in patients with poor baseline visual acuity.

OBJECTIVE: To report the visual outcomes of intravitreal anti-vascular endothelial growth factor (anti-VEGF) treatment for macular edema secondary to central retinal vein occlusion (CRVO) in patients with baseline visual acuity of ≤23 ETDRS letters vision. DESIGN: Retrospective observational cohort study. METHODS: This is a single-institution study. A total of 173 eyes from 173 patients who had completed 3 consecutive monthly anti-VEGF injections for macular edema secondary CRVO and best-corrected visual acuity (BCVA) ≤23 ETDRS letters were included. The main outcome measures were visual acuity at month 3 and month 12. RESULTS: At month 3, BCVA increased to 34.1 ETDRS letters (95% CI, 30.7-37.5), with a gain of 25.0 letters (95% CI, 22.0-28.5; p < 0.001). The mean central subfield thickness decreased by 519 µm (95% CI, 475.5-567.0; p < 0.001). Most patients (67.6%) gained >15 ETDRS letters. A total of 160 patients were followed up for 12 months, and the mean BCVA was 31.2 ETDRS letters (95% CI, 27.5-34.9) at the end of this period. A third of eyes that did not respond (<5-letter gain) after a single injection experienced a 15-letter or more improvement after 3 consecutive injections. CONCLUSIONS: Anti-VEGF treatment in eyes with CRVO and poor baseline visual acuity results in significant visual improvement, and moderate improvement is still noted despite a poor response after a single injection.

Choroidal Involvement in a Case of Viral Retinitis.

PURPOSE: To describe a challenging case of acute retinal necrosis. METHODS: Clinical data, including medical history, clinical findings and imaging features were retrospectively collected in a 28-year-old female with unilateral active retinitis. RESULTS: The yellowish area of retinitis surrounding a pigmented chorioretinal scar and the full thickness retinal hyper-reflectivity associated with diffuse increase in choroidal demonstrated a challenging case of necrotising retinitis, where baseline clinical and tomographic features were atypical and misleading towards a toxoplasmic aetiology. The detection of virus genome in ocular samples was necessary to achieve a correct diagnosis of acute retinal necrosis. CONCLUSION: Although rare, choroidal involvement may occur in acute retinal necrosis along with the well-known retinal features.

Inter-rater reliability for diagnosis of geographic atrophy using spectral domain OCT in age-related macular degeneration.

PURPOSE: To evaluate the inter-rater reliability for identification of complete retinal pigment epithelium and outer retinal atrophy (cRORA) on SD-OCT images as defined by the Classification of Atrophy Meetings (CAM) group. METHODS: Fifty images of anonymized SD-OCT line scans of eyes with cRORA due to AMD were selected. Each .tiff image was saved in both black-on-white (BW) and white-on-black (WB) format. Five retina-trained clinicians graded both sets of images twice for the diagnosis of cRORA based on the CAM group definition. Fleiss kappa statistic was calculated for inter-rater reliability and Cohen's kappa statistic for intra-grader and inter-grader reliability between any two graders. RESULTS: The inter-grader reliability varied from as low as 0.28 to 0.92 for WB images and 0.34 to 0.86 for BW images. However, the inter-grader and intra-grader agreement was ĸ WB 0.92; ĸ BW 0.86 and ĸ 0.92 respectively, for graders accustomed to the CAM criteria. Fleiss kappa was ĸ 0.49 (p value < 0.0001) for WB images and ĸ 0.34 (p value < 0.0001 for BW images. Overall, the agreement was better using WB images for all parameters except RPE attenuation/loss. CONCLUSION: There is significant variability in diagnosis of cRORA on SD-OCT by retina-trained ophthalmologists in the real world. The study highlights the need for training to recognise the different features of cRORA prior to its implementation in clinical practice.

Functional clinical endpoints and their correlations in eyes with AMD with and without subretinal drusenoid deposits-a pilot study.

PURPOSE: To evaluate functional clinical endpoints and their structural correlations in AMD, with a focus on subretinal drusenoid deposits (SDD). METHODS: This prospective study enroled 50 participants (11 controls, 17 intermediate AMD (iAMD) with no SDD, 11 iAMD with SDD and 11 non-foveal atrophic AMD). Participants underwent best-corrected visual acuity (BCVA), low luminance visual acuity (LLVA), low luminance questionnaire (LLQ), scotopic thresholds, rod-intercept time (RIT), photopic flicker electroretinograms and multimodal imaging. Functional and structural relationships were assessed. RESULTS: Compared with healthy participants, BCVA, LLVA, scotopic thresholds were depressed, and RIT prolonged in iAMD patients with SDD (p = 0.028, p = 0.045, p = 0.014 and p < 0.0001 respectively). Patients with SDD also had reduced scotopic function and delayed RIT compared to iAMD without SDD (p = 0.005 and p < 0.0001). Eyes with SDD and non-foveal atrophy did not differ functionally. Nor did healthy subjects compared with iAMD without SDD. Functional parameters were significantly associated with scotopic thresholds (r = 0.39-0.64). BCVA, LLVA and scotopic thresholds correlated well with ONL volume, ONL thickness and choroidal thickness (r = 0.34-0.61). CONCLUSION: Eyes with SDD are surrogate markers of photoreceptor abnormalities comparable with non-central atrophy and should be sub-analysed in clinical trials evaluating potential prophylactic agents to decrease the progression of AMD and may even require different therapeutic interventions.

The Diagnostic Accuracy of Double-Layer Sign in Detection of Macular Neovascularization Secondary to Central Serous Chorioretinopathy.

PURPOSE: To investigate the diagnostic value of elevated retinal pigment epithelium (RPE) and double-layer sign (DLS) in identifying macular neovascularization (MNV) secondary to central serous chorioretinopathy (CSCR). DESIGN: Retrospective, cross-sectional study. METHODS: Patients with CSCR underwent optical coherence tomography (OCT) and OCT angiography (OCT-A) scanning at Moorfields Eye Hospital. OCT scans were reviewed to identify the presence/absence of an RPE elevation. The maximum length and maximum height of the elevated RPE were measured. A minimum length of 1000 µm and a maximum height of 150 µm were used to define the "double-layer sign." Other qualitative anatomical features were also graded from OCT scans. OCT-A was examined to confirm the presence/absence of MNV. Binary logistic regression analyses were used to assess the association between OCT features and the detection of MNV on OCT-A. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated to assess the diagnostic accuracy. RESULTS: One hundred sixty-three eyes from 132 patients were included. Elevated RPE was detected in 148 eyes (91%). OCT-A-confirmed MNV was detected in 54 eyes (33%). The sensitivity and specificity of RPE elevation were 100% and 13.8%, respectively. DLS was identified in 95 eyes (58%). The sensitivity and specificity of DLS for detecting MNV were 87% and 56%, respectively. Hyperreflectivity and nonhomogeneity of the sub-RPE space were independently associated with MNV within the DLS (odds ratio, 17.7 and 14.8, P < .001 and P = .02, respectively). None of the other demographic or anatomical features were associated with MNV. The presence of nonhomogeneous hyperreflective RPE elevation had a sensitivity and specificity of 98% and 67%, with PPV and NPV of 60% and 99%, respectively. CONCLUSIONS: Nonhomogeneous and hyperreflective space under an elevated RPE of any length or height indicates an eye with higher risk of MNV than DLS. OCT-A should at least be performed for these eyes to confirm the presence of MNV and treat accordingly.

Deep learning for gradability classification of handheld, non-mydriatic retinal images.

Screening effectively identifies patients at risk of sight-threatening diabetic retinopathy (STDR) when retinal images are captured through dilated pupils. Pharmacological mydriasis is not logistically feasible in non-clinical, community DR screening, where acquiring gradable retinal images using handheld devices exhibits high technical failure rates, reducing STDR detection. Deep learning (DL) based gradability predictions at acquisition could prompt device operators to recapture insufficient quality images, increasing gradable image proportions and consequently STDR detection. Non-mydriatic retinal images were captured as part of SMART India, a cross-sectional, multi-site, community-based, house-to-house DR screening study between August 2018 and December 2019 using the Zeiss Visuscout 100 handheld camera. From 18,277 patient eyes (40,126 images), 16,170 patient eyes (35,319 images) were eligible and 3261 retinal images (1490 patient eyes) were sampled then labelled by two ophthalmologists. Compact DL model area under the receiver operator characteristic curve was 0.93 (0.01) following five-fold cross-validation. Compact DL model agreement (Kappa) were 0.58, 0.69 and 0.69 for high specificity, balanced sensitivity/specificity and high sensitivity operating points compared to an inter-grader agreement of 0.59. Compact DL gradability model performance was favourable compared to ophthalmologists. Compact DL models can effectively classify non-mydriatic, handheld retinal image gradability with potential applications within community-based DR screening.

Predictors of Visual Acuity Outcomes after Anti-Vascular Endothelial Growth Factor Treatment for Macular Edema Secondary to Central Retinal Vein Occlusion.

PURPOSE: To evaluate whether baseline demographic, clinical, and OCT characteristics predict visual acuity (VA) outcomes in patients receiving anti-vascular endothelial growth factor (VEGF) therapy for macular edema (ME) due to central retinal vein occlusion (CRVO). DESIGN: Post hoc analysis of the randomized noninferiority trial (Lucentis, Eylea, Avastin in CRVO) LEAVO Study from December 12, 2014, to December 16, 2016, carried out across 44 UK National Health Service ophthalmology departments. PARTICIPANTS: Data on 267 participants with a baseline best-corrected mean visual acuity (BCVA) range of 19 to 78 Early Treatment Diabetic Retinopathy Study letter score (approximate Snellen equivalent, 20/32 to 20/320) who had central subfield thickness (CST) ≥ 320 µm on Spectralis OCT (Heidelberg Engineering) were analyzed. METHODS: Study participants were randomized to receive repeated intravitreal injections of ranibizumab (0.5 mg/50 µl), aflibercept (2.0 mg/50 µl), or bevacizumab (1.25 mg/50 µl), and a protocol-driven pro re nata re-treatment regimen at 4 to 8 weekly visits was followed up to week 100 after 4 mandated 4-weekly loading injections. MAIN OUTCOME MEASURES: Change in BCVA and percentage of patients gaining ≥ 10 letters and achieving BCVA letter score > 70 letters at 52 and 100 weeks. RESULTS: The analysis was adjusted for treatment effects and confirmed by sensitivity analysis. Age ≥ 75 years is a poor predictor for all 3 visual outcomes. Lower baseline BCVA predicted 10-letter gainers and higher gains in BCVA, although it is a poor predictor of achieving > 70 Early Treatment Diabetic Retinopathy Study letters. None of the baseline OCT morphologic characteristics except ellipsoid zone (EZ) integrity influenced any visual outcomes. Both baseline CST and total macular volume showed a nonlinear relation to 10-letter gainers, with CST > 900 µm being a poor prognostic indicator. Baseline CST and macular volume did not predict mean change in BCVA or BCVA > 70 letters at 52 and 100 weeks. The sensitivity analysis conclusions after removing iCRVO were similar. CONCLUSIONS: At presentation, younger age, higher baseline BCVA, and a definitely intact subfoveal EZ are predictors of BCVA score > 70 letters at 100 weeks.

Impact of injection frequency on 5-year real-world visual acuity outcomes of aflibercept therapy for neovascular age-related macular degeneration.

BACKGROUND: To evaluate the impact of injection frequency on yearly visual outcomes of patients treated with intravitreal aflibercept for neovascular age-related macular degeneration (nAMD) over a period of 5 years in a tertiary ophthalmic centre. DESIGN: Single centre, retrospective cohort study. PARTICIPANTS: Consecutive treatment-naive nAMD patients initiated on aflibercept injections 5 years ago. METHODS: The Moorfields OpenEyes database was searched for consecutive patients who were initiated on intravitreal aflibercept for nAMD in 2013-14 and the visual acuity (VA) in Early Diabetic Retinopathy Study (ETDRS) letters and injection records per year were recorded for a period of 5 years. Analyses of the whole cohort and a sub-sample of 5-year completers were done. The cohort was further grouped into Group A (on continuous treatment), Group B (early cessation of treatment) and Group C (interrupted treatment) to evaluate the relation between treatment frequency and visual outcomes. MAIN OUTCOME MEASURES: The primary end point was change in VA at 5 years; secondary outcomes included proportion of eyes that gained or maintained VA, number of injections received and the effect of treatment frequency. RESULTS: Data were collected on 468 patients (512 eyes). Sixty-six percent of the patients completed 5-year follow-up. The mean age of the whole cohort was 79.5 ± 8.5 years and the mean baseline VA was 58.3 ± 15.4 letters. Amongst the completers, final VA change was -2.9 (SD 23.4) ETDRS letters and the cumulative number of injections over 5 years was 24.2 (10.6). Group A had three letter gain and received significantly higher cumulative number of injections over 5 years than Group B and C (31.8, 14.6 and 18.4 respectively, p = 0.001). After adjusting for age and baseline VA, on average, final VA was +8.0 letters higher in the ≥20 injections group than the <20 group (p = 0.001). CONCLUSIONS: Aflibercept therapy results in sustained good visual outcome over 5 years in neovascular AMD eyes when early and persistent treatment is given.

Deep Learning-Based Segmentation and Quantification of Retinal Capillary Non-Perfusion on Ultra-Wide-Field Retinal Fluorescein Angiography.

Reliable outcome measures are required for clinical trials investigating novel agents for preventing progression of capillary non-perfusion (CNP) in retinal vascular diseases. Currently, accurate quantification of topographical distribution of CNP on ultrawide field fluorescein angiography (UWF-FA) by retinal experts is subjective and lack standardisation. A U-net style network was trained to extract a dense segmentation of CNP from a newly created dataset of 75 UWF-FA images. A subset of 20 images was also segmented by a second expert grader for inter-grader reliability evaluation. Further, a circular grid centred on the FAZ was used to provide standardised CNP distribution analysis. The model for dense segmentation was five-fold cross-validated achieving area under the receiving operating characteristic of 0.82 (0.03) and area under precision-recall curve 0.73 (0.05). Inter-grader assessment on the 20 image subset achieves: precision 59.34 (10.92), recall 76.99 (12.5), and dice similarity coefficient (DSC) 65.51 (4.91), and the centred operating point of the automated model reached: precision 64.41 (13.66), recall 70.02 (16.2), and DSC 66.09 (13.32). Agreement of CNP grid assessment reached: Kappa 0.55 (0.03), perfused intraclass correlation (ICC) 0.89 (0.77, 0.93), non-perfused ICC 0.86 (0.73, 0.92), inter-grader agreement of CNP grid assessment values are Kappa 0.43 (0.03), perfused ICC 0.70 (0.48, 0.83), non-perfused ICC 0.71 (0.48, 0.83). Automated dense segmentation of CNP in UWF-FA images achieves performance levels comparable to inter-grader agreement values. A grid placed on the deep learning-based automatic segmentation of CNP generates a reliable and quantifiable method of measurement of CNP, to overcome the subjectivity of human graders.