PURPOSE: To describe the clinical course and treatment outcomes of inflammatory choroidal neovascular membranes (CNVM) in patients with Vogt-Koyanagi-Harada (VKH) disease. METHODS: This is a retrospective chart review wh...PURPOSE: To describe the clinical course and treatment outcomes of inflammatory choroidal neovascular membranes (CNVM) in patients with Vogt-Koyanagi-Harada (VKH) disease. METHODS: This is a retrospective chart review which included all VKH patients who were diagnosed with inflammatory CNVM at King Khaled Eye Specialist Hospital (KKESH) between the years 2014 and 2025. RESULTS: Fifty-seven eyes of 37 VKH patients were diagnosed with CNVM. The mean age of the patients at presentation was 35.3 ± 14.6 years. The type of VKH was chronic recurrent in 25 patients (67.6%), while it was initial-onset acute in 12 patients (32.4%). In 42 eyes (73.7%), the CNVM was present at presentation, while 15 eyes (26.3%) developed CNVM after an average of 37.3 ± 29.4 months. The CNVM was bilateral in 20 patients (54.1%) while 17 patients (45.9%) had unilateral disease. Treatments were given to 49 eyes (85.9%), where anti-VEGF was given to 47 (82.5%) and photodynamic therapy to 2 (3.5%). Multiple anti-VEGF agents were used in 20 eyes (41.7%). The mean number of CNVM reactivations was 3.98 ± 3.9 times in 43 eyes (75.4%). The mean number of injections was 8.4 ± 10.6 injections. There was an overall significant improvement in BCVA at the last follow-up visit (p=0.008). The reduction in central retinal thickness was statistically significant (p<0.001). CONCLUSION: Inflammatory CNVM is an infrequent complication in patients with VKH. Good visual and anatomical improvements can be achieved with appropriate control of systemic inflammation and timely management of the CNVM.
PURPOSE: Inflammatory cytokines are key mediators of diabetic macular edema (DME). Prior assessments are limited by study design and small sample size. We rigorously analyzed the association of 24 inflammatory cytokines...PURPOSE: Inflammatory cytokines are key mediators of diabetic macular edema (DME). Prior assessments are limited by study design and small sample size. We rigorously analyzed the association of 24 inflammatory cytokines with DME in a prospective trial of 328 eyes of 164 adult type II diabetic patients with different stages of diabetic retinopathy severity. METHODS: ETDRS visual acuity, fundus photography, and OCT of each eye were taken at baseline. Bilateral aqueous sampling was performed. A microparticle bead-based multiplex assay was used to measure aqueous levels of 24 inflammatory cytokines. DME was defined as central subfield thickness > 320 µm on OCT with the presence of intraretinal fluid. RESULTS: Median and interquartile range (IQR) of VEGF-A, IL-6, and IL-8 were 148.70 pg/ml (103.93-232.96), 7.67 pg/ml (4.64-18.90), and 11.14 pg/ml (6.48-17.46) in eyes without DME, respectively and 181.32 pg/ml (141.82-307.81), 16.53 pg/ml (8.15-48.68), and 15.95 (10.67-26.53) in eyes with DME, respectively. Positive odds ratios with 95% confidence intervals were observed for IL-8 (2.73 [1.41-5.29]; P=0.003), FLT-3L (2.27 [1.22-4.23]; P=0.01), FGF-2 (1.74 [1.06-2.86]; P=0.029), IL-6 (2.93 [1.62-5.31]; P<0.001), and IL-10 (2.79 [1.50-5.21]; P=0.001). CONCLUSIONS: We report that IL-8, FLT-3L, FGF-2, IL-6, and IL-10were significantly elevated in eyes with DME.
PURPOSE: To characterize outer retina segmentation slab (ORSS) vascularization in Type 2 macular telangiectasia (MacTel) using multimodal imaging and to develop a regression-derived scoring system for differentiating tru...PURPOSE: To characterize outer retina segmentation slab (ORSS) vascularization in Type 2 macular telangiectasia (MacTel) using multimodal imaging and to develop a regression-derived scoring system for differentiating true neovascularization from non-neovascular vascularization. METHODS: In this retrospective study, 336 eyes of 174 patients with Type 2 MacTel were screened, of which 150 eyes (80 patients) with analysable optical coherence tomography angiography (OCTA) were included. ORSS vascularization was identified in 92 eyes (61.3%). Multimodal imaging features from optical coherence tomography (OCT), fundus fluorescein angiography (FFA), and OCTA were analyzed. Multivariable logistic regression was performed to identify predictors of neovascularization (subretinal neovascular membrane; SRNVM), and a weighted composite score (ORSS-V score) was developed and validated using receiver operating characteristic (ROC) analysis. RESULTS: Among eyes with ORSS vascularization, 33 (35.9%) demonstrated SRNVM, while 59 (64.1%) did not. Eyes with SRNVM were older (mean: 63.4 years) and had worse visual acuity [Mean logMAR = 0.62; 20/83] (p ≤ 0.001). Key differentiating features included subretinal and intraretinal fluid on OCT, focal intense leakage on FFA, complex OCTA vascular network, and deep capillary plexus vascular bunching (all p < 0.001). The ORSS-V score demonstrated excellent discriminative ability (area under ROC = 0.91). At a cut-off ≥6, sensitivity was 88% and specificity 86%. CONCLUSION: ORSS vascularization is common in MacTel and often non-neovascular. A multimodal imaging approach, supported by the ORSS-V scoring system, enables reliable differentiation between invasive vascularization and true neovascularization, improving diagnostic accuracy and guiding clinical management.
PURPOSE: To evaluate the long-term effects of preoperative ectopic inner foveal layer (EIFL) with preserved foveal contour on postoperative retinal microvascular changes in patients with epiretinal membrane. METHODS: Pat...PURPOSE: To evaluate the long-term effects of preoperative ectopic inner foveal layer (EIFL) with preserved foveal contour on postoperative retinal microvascular changes in patients with epiretinal membrane. METHODS: Patients were divided into two groups based on EIFL presence: Group 1 without EIFL and Group 2 with EIFL. Superficial and deep vessel densities (SVD and DVD), corresponding to the superficial and deep vascular complexes, were quantitatively assessed using OCT angiography. RESULTS: The SVD at baseline, 1, 3, 6, and 12 months were 27.9 ± 7.7, 27.6 ± 8.3, 25.9 ± 7.8, 26.1 ± 7.0, and 24.9 ± 8.0 % in Group 1, respectively (P = 0.075), and 31.6 ± 7.3, 28.0 ± 6.3, 26.7 ± 5.6, 25.4 ± 5.7, and 22.9 ± 5.6 % in Group 2, respectively (P < 0.001). The DVDs were 21.9 ± 4.6, 22.9 ± 4.5, 23.6 ± 5.1, 24.2 ± 4.0, and 23.6 ± 4.7 % in Group 1, respectively (P = 0.277), and 20.6 ± 4.7, 22.8 ± 4.2, 23.2 ± 3.7, 23.6 ± 5.0, and 23.4 ± 3.8 % in Group 2, respectively (P = 0.003). In the multivariable analysis, baseline BCVA was significantly associated with changes in BCVA in Group 1, while in Group 2, baseline BCVA and DVD showed significant associations with BCVA changes. CONCLUSIONS: Patients with EIFL and preserved foveal contour demonstrated more prominent alterations in retinal microvasculature after surgery, and these changes, especially within the deep vascular complex, are considered to play an important role in visual improvement.
PURPOSE: To analyze macular hole development using three-dimensional visualization of vitreomacular architecture volume-rendered optical coherence tomography (OCT) and to interpret observed anatomic configurations using...PURPOSE: To analyze macular hole development using three-dimensional visualization of vitreomacular architecture volume-rendered optical coherence tomography (OCT) and to interpret observed anatomic configurations using structural and mechanical principles. METHODS: This retrospective observational study included eyes with developing macular holes imaged using a swept-source OCT. Volumetric datasets of the macula and overlying vitreous were processed with extended dynamic-range processing and rendered to permit three-dimensional visualization of vitreous attachment, deformation, and retinal structural changes. Anatomic configurations were evaluated descriptively. Terminology was applied to denote OCT-based structural failure states without implying a fixed temporal sequence. RESULTS: Seventeen eyes of thirteen patients were analyzed. Persistent vitreous attachment to the central macula was associated with a range of distinct anatomic configurations, including inner macular cavitation, combined inner cavitation with outer retinal separation, outer macular holes with preserved inner retina, inner lamellar holes, and full-thickness macular holes. Volume rendering demonstrated asymmetric vitreous attachment, catenary-like vitreous geometry, tears in the posterior cortical vitreous, vitreous prolapse, and frequent retinal flaps that were not readily apparent on conventional planar imaging. Longitudinal imaging showed variable transitions between structural configurations associated with changes in vitreous attachment rather than a uniform progression through predefined stages. CONCLUSIONS: Volume-rendered swept-source OCT enables direct visualization of vitreomacular architecture during macular hole development. The observed anatomic configurations are best understood as traction-induced structural failure states of the central macula rather than as a linear temporal staging sequence. This framework better accounts for the diversity, asymmetry, and nonuniform evolution of macular hole morphology observed in vivo.
PURPOSE: To investigate whether the morphology of supra-retinal pigment epithelial (RPE) deposits influences anatomical and functional outcomes in idiopathic full-thickness macular holes (FTMHs). METHODS: In this retrosp...PURPOSE: To investigate whether the morphology of supra-retinal pigment epithelial (RPE) deposits influences anatomical and functional outcomes in idiopathic full-thickness macular holes (FTMHs). METHODS: In this retrospective study 66 eyes from 66 patients with idiopathic FTMH undergoing vitrectomy with ILM peeling were analyzed. Supra-RPE deposits were segmented from en face OCT slabs. Deposits were classified as fine deposits (FD) or granular deposits (GD) based on size. Baseline MH characteristics were measured on horizontal and vertical B-scans. Outcomes were assessed six months after surgery. Best-corrected visual acuity (VA) was recorded in logMAR units. RESULTS: 38 eyes (58%) showed FD and 28 eyes (42%) GD. Baseline characteristics were comparable between groups. At 6 months after surgery, eyes with GD achieved worse postoperative VA (p=0.03) and showed greater disruption of outer retinal bands, with larger EZd (p=0.035) and ELMd (p=0.048). ΔVA was significantly smaller in GD eyes (-0.40 vs. -0.65 logMAR, p<0.001). In FD eyes deposit area correlated with postoperative changes in EZd and ELMd, whereas no such association was found in GD eyes. In multivariable regression deposit morphology (GD vs. FD) independently predicted poorer visual improvement (β=0.15, 95% CI [0.02-0.29], p=0.03). CONCLUSION: Supra-RPE deposits are frequently detectable by en face OCT in FTMH. GD identifies eyes with larger MHs and reduced likelihood of optimal postoperative functional recovery. These findings suggest that deposit phenotype may serve as an imaging biomarker of disease severity, though further work is required to confirm its independent prognostic value.
PURPOSE: To evaluate the safety and efficacy of a ring indenter scleral self-indentation technique for basal vitrectomy under noncontact wide-angle viewing in primary rhegmatogenous retinal detachment (RRD) repair compar...PURPOSE: To evaluate the safety and efficacy of a ring indenter scleral self-indentation technique for basal vitrectomy under noncontact wide-angle viewing in primary rhegmatogenous retinal detachment (RRD) repair compared with conventional assistant-assisted indentation. METHODS: This single-center retrospective cohort study enrolled 99 eyes of 99 primary RRD patients who underwent pars plana vitrectomy with basal vitreous shaving from August 2023 to May 2025, with ≥6 months follow-up. Patients were divided into study group (52 eyes, surgeon's finger ring scleral self-indentation) and control group (47 eyes, conventional assistant-assisted indentation). Main outcomes: retinal redetachment rate, final BCVA, and shaving-induced retinal breaks (SIRBs) incidence. RESULTS: The baseline demographic and clinical characteristics were comparable between the two groups. The retinal redetachment rate was 1.92% (1/52) in the study group and 2.13% (1/47) in the control group (P=1.000). Both groups demonstrated significant improvements in BCVA after the surgery compared with the preoperative values (P<0.0001). No significant differences were observed between the two groups in the final BCVA (P=0.365) and BCVA improvement (P=0.168). The incidence of SIRBs was one case in each group (P=1.000). No intraoperative instrument displacement, infusion cannula dislodgement, or anesthesia-related complications occurred in either group. CONCLUSIONS: The ring indenter scleral self-indentation technique is a safe and effective alternative to assistant-assisted indentation for basal vitreous clearance during RRD vitrectomy. This cost-effective technique eliminates the need for assistant support or chandelier illumination, offering the advantages of autonomous and precise manipulation, clear visualization, and manpower conservation.
PURPOSE: To determine factors associated with persistent vision loss among eyes with nonproliferative diabetic retinopathy (NPDR) and diabetic macular edema (DME) who return to care following loss to follow-up (LTFU). ME...PURPOSE: To determine factors associated with persistent vision loss among eyes with nonproliferative diabetic retinopathy (NPDR) and diabetic macular edema (DME) who return to care following loss to follow-up (LTFU). METHODS: This retrospective study was conducted at the Wills Eye Hospital/Mid Atlantic Retina between January 1st, 2015 and January 1st, 2023. Patients were included if they met the following study inclusion criteria I) ≥1 anti-VEGF injection before LTFU, II) history of LTFU for >6 months and, III) returned for care with ≥12 months of follow-up after LTFU. RESULTS: A total of 169 eyes of 169 patients met our inclusion criteria. Approximately 38% of eyes (N=64/169) worsened after returning to care, with ≥10-letter decline in VA compared to their pre-LTFU visit. At 12 months post-return, 27.2% (N=46/169) had a persistent ≥10-letter decrease in VA from their pre-LTFU baseline. On multivariable analysis, white race (OR 0.42, 95% CI: 0.20-0.92, p=0.03) was associated with decreased likelihood of ≥10 letter loss at the 12-month follow-up visit. VA at time of return (OR 6.47, 95% CI: 1.96-21.33, p=0.002) was also associated with vision loss at the 12-month visit. Increasing CFT at time of return (OR 1.004, 95% CI: 1.001-1.007, p=0.012) was the only factor associated with a ≥10 letter loss at the return visit. CONCLUSION: Approximately 1/4th of eyes had persistent vision loss at 1-year despite resumption of treatment. Race and VA upon return following LTFU were associated with vision loss at the 12-month visit. Increased macular edema was associated with vision loss at return visit.
PURPOSE: To evaluate 12-month outcomes following systematic switch to aflibercept 8 mg in previously treated eyes with age-related neovascular macular degeneration (nAMD). METHODS: Eyes in a plateau phase (≥10 prior anti...PURPOSE: To evaluate 12-month outcomes following systematic switch to aflibercept 8 mg in previously treated eyes with age-related neovascular macular degeneration (nAMD). METHODS: Eyes in a plateau phase (≥10 prior anti-VEGF injections) that were switched to aflibercept 8 mg were retrospectively evaluated. Treatment intervals, best corrected visual acuity (BCVA), central retinal thickness (CRT), and data on intraretinal fluid (IRF) or subretinal fluid (IRF) were assessed at switch and after 3, 6, and 12 months. RESULTS: In total, 654 eyes from 567 patients were included (mean [SD] age 82.4 [7.4], 34.5 [21.5] injections per eye before switch). The proportion of eyes with dry maculae increased from 55% at switch to 76% at 3 months (P<0.001), 73% at 6 months (P<0.001), and 69% at 12 months (P<0.05). Eyes treated at 3-5-week intervals declined from 46% to 18% at 12 months (P<0.001). Eyes on 9-12-week intervals increased from 11% to 34% (P<0.001). The mean treatment interval increased from 6.0 (2.1) to 7.8 (2.4) weeks (P<0.001). Eyes with dry maculae at switch were extended by 2.6 (1.8) weeks (P<0.001). The mean CRT decreased from 234.7 (49.7) to 223.5 (46.7) µm (P<0.001). BCVA declined from 68.5 (12.2) to 67.3 (13.9) ETDRS letters in the overall cohort, and from 68.5 (12.3) to 66.6 (14.3) ETDRS letters in eyes with a dry macula at switch (P < 0.001). CONCLUSION: Switching to aflibercept 8 mg improved anatomical outcomes and reduced the proportion of eyes on short intervals. A slight decline in BCVA was observed, warranting long-term follow-up.
PURPOSE: To determine whether aqueous flare reflects intraocular inflammation in treatment-naïve diabetic macular edema (DME) and to examine associations between aqueous cytokines and optical coherence tomography (OCT) b...PURPOSE: To determine whether aqueous flare reflects intraocular inflammation in treatment-naïve diabetic macular edema (DME) and to examine associations between aqueous cytokines and optical coherence tomography (OCT) biomarkers, using Irvine-Gass syndrome (IGS) and cataract eyes as controls. METHODS: Seventy-six eyes were included (40 DME, 19 IGS, 17 cataract controls). All eyes underwent OCT imaging and laser flare photometry. Aqueous levels of interleukin-6 (IL-6), monocyte chemoattractant protein-1 (MCP-1), interferon-γ-inducible protein-10 (IP-10), and vascular endothelial growth factor (VEGF) were measured. OCT biomarkers included hyperreflective foci (HRF), ellipsoid zone (EZ) disruption, and subretinal fluid. RESULTS: IL-6 levels differed among groups (p = 0.015), being higher in IGS than controls, while MCP-1 showed a borderline difference (p = 0.054) with higher levels in IGS versus DME (p = 0.008). IP-10 did not differ among groups, and VEGF levels were similar between IGS and DME (p = 0.659). Flare values were higher in IGS than controls (p = 0.039) but did not differ between DME and controls. Compared with IGS, DME eyes more frequently exhibited HRF and EZ disruption (all p <0.05). In DME, HRF count correlated with IP-10 on univariate analysis (r = 0.33, p <0.05), but this association did not remain significant after multivariate adjustment (p = 0.075). CONCLUSION: In treatment-naïve DME eyes, inflammatory profiles likely reflect an early disease state and may differ from those observed in chronic or previously treated DME. HRF may represent an OCT phenotype related to inflammatory signaling, although an independent association with IP-10 was not confirmed, with potential relevance for future studies.
PURPOSE: To evaluate plasma and aqueous humor angiotensin-converting enzyme (ACE) and transforming growth factor-β1 (TGFβ1) levels in patients with proliferative vitreoretinopathy (PVR) secondary to rhegmatogenous retina...PURPOSE: To evaluate plasma and aqueous humor angiotensin-converting enzyme (ACE) and transforming growth factor-β1 (TGFβ1) levels in patients with proliferative vitreoretinopathy (PVR) secondary to rhegmatogenous retinal detachment (RRD). METHODS: Eighty-eight patients were divided into three groups: RRD without PVR (non-PVR, n = 36), RRD with PVR (PVR, n = 32), and controls (n = 20). Plasma and aqueous ACE and TGFβ1 levels were quantified using ELISA. Group comparisons, regression, and ROC analyses were performed to determine associations with PVR. RESULTS: Age and sex did not differ among groups (p > 0.05). Median aqueous ACE levels were significantly higher in PVR and non-PVR eyes than in controls (both p < 0.001), though the PVR-non-PVR difference was nonsignificant (p = 0.411). Aqueous TGFβ1 levels were markedly elevated in the PVR group versus both others (p < 0.001). Plasma ACE and TGFβ1 did not differ among groups (p > 0.05). Aqueous TGFβ1 demonstrated strong discriminative performance for PVR within this cohort (AUC = 0.92; cutoff 132.0 ng/L; sensitivity 93.8%; specificity 72.2%). Multivariate analysis identified aqueous TGFβ1 (OR = 1.047, p = 0.002) and symptom duration >30 days (OR = 77.6, p = 0.014) as independent predictors. Aqueous ACE and TGFβ1 were positively correlated, although this association did not reach statistical significance (ρ = 0.230, p = 0.059). CONCLUSION: Aqueous ACE and TGFβ1 levels were elevated in eyes with RRD. TGFβ1 was significantly associated with PVR, whereas ACE elevation appeared to reflect intraocular changes related to RRD rather than a PVR-specific process.
PURPOSE: To evaluate clinical factors that lead to ≥15-letter visual acuity loss following plaque radiotherapy of posterior uveal melanoma (PUM) in the pre-anti-vascular endothelial growth factor (anti-VEGF) era. METHODS...PURPOSE: To evaluate clinical factors that lead to ≥15-letter visual acuity loss following plaque radiotherapy of posterior uveal melanoma (PUM) in the pre-anti-vascular endothelial growth factor (anti-VEGF) era. METHODS: We reviewed data on 4000 consecutive patients with plaque-irradiated PUM managed between May 1976 and November 2008 to identify parameters predictive of ≥15-letter vision loss. RESULTS: Data included median patient age (60 years), diabetes mellitus (11%), median tumor diameter (11.0 mm) and thickness (4.5 mm), and median distance to optic disc (4.0 mm) and foveola (3.0 mm). Kaplan-Meier analyses revealed ≥15-letter loss at 1 versus (vs.) 3 vs. 5 vs. 10 years in 30% vs. 60% vs. 73% vs. 82% of patients. Multivariate analyses revealed factors predictive of ≥15-letter loss included greater tumor thickness (p<0.01, hazard ratio [HR] 1.14), shorter distance to foveola (p<0.01, HR 0.93), greater radiation dose to foveola (p<0.01, HR 1.01) and lens (p<0.01, HR 1.01), and presence of diabetes mellitus (p<0.01, HR 1.11). For small PUM, multivariate risk factors included older age (p<0.01), diabetes mellitus (p<0.01), and shorter distance to foveola (p<0.02); for medium PUM, risk factors included greater tumor thickness (p<0.01), shorter distance to foveola (p<0.01), and less (but shorter) radiation dose to foveola (p<0.01); for large PUM, risk factors included greater tumor thickness (p<0.01) only. CONCLUSIONS: In the pre-anti-VEGF days, following treatment of PUM with plaque radiotherapy, clinical and treatment factors predictive of ≥15-letter visual acuity loss included greater tumor thickness, shorter distance to foveola, greater radiation dose to foveola and lens, and presence of diabetes mellitus.
PURPOSE: To evaluate independent pre-operative risk factors for secondary macular hole (MH) following fovea-off rhegmatogenous retinal detachment (RRD) and their impact on post-operative visual outcomes. METHODS: This re...PURPOSE: To evaluate independent pre-operative risk factors for secondary macular hole (MH) following fovea-off rhegmatogenous retinal detachment (RRD) and their impact on post-operative visual outcomes. METHODS: This retrospective study included fovea-off RRD patients with pre-operative optical coherence tomography (OCT) confirmed sub-foveal fluid and complete pre- and post-operative demographics, clinical features, and OCT images. Based on post-operative OCT images, patients were subdivided into No MH, lamellar MH, and full-thickness MH groups after RRD surgery. Multivariate logistic regression and linear regression were applied to identify the factors associated with post-operative secondary MH and best-corrected visual acuity (BCVA) respectively. RESULTS: Of these 705 patients, 57 developed secondary MH after RRD surgery, while the remaining 648 patients served as the control group. Multivariate logistic regression analysis demonstrated that Müller cell cone (MCC) loss (OR = 4.401, P = 0.001), bacillary layer detachment (OR = 2.528, P = 0.041), retinal detachment (RD) extent (OR = 1.136, P = 0.031), and foveal neurosensory thickness (OR = 0.992, P = 0.030) were independent MH risk factors. Multiple linear regression analysis showed that MCC loss (β = 0.308, P < 0.001), greater RD extent (β = 0.024, P = 0.010), longer RD duration (β = 0.001, P = 0.036), outer nuclear layer thinning (β = 0.144, P = 0.001), and worse pre-operative BCVA (β = 0.129, P < 0.001) predicted poorer visual outcomes. CONCLUSION: MCC loss is the strongest risk factor for both MH development and worse visual prognosis after RRD surgery.