The relationship between coronavirus disease 2019 (COVID-19) infection or vaccination and retinal artery occlusions (RAOs) remains controversial. COVID-19 infection or vaccination can sometimes cause thrombin formation. RAOs occur due to thrombin in the retinal artery. The average age of occurrence after COVID-19 infection was 48.7 ± 17.2 years in central RAO (CRAO) and 41.3 ± 17.8 years in branch RAO (BRAO). After COVID-19 vaccination, the average age was 54.7 ± 17.1 years in CRAO and 62.3 ± 21.2 years in BRAO. The mean time from COVID-19 diagnosis to symptom onset was 10.5 ± 9.3 days in CRAO and 48.3 ± 39.6 days in BRAO. After vaccination, the mean time was 7.3 ± 6.8 days in CRAO and 21.0 ± 24.9 days in BRAO. Initial visual acuity (VA) after COVID-19 infection was 2.53 ± 0.65 in CRAO and 0.09 ± 0.07 in BRAO. Final VA was 2.73 ± 0.12 in CRAO, but data for BRAO were unavailable. After vaccination, initial VA was 2.28 ± 0.97 in CRAO and -0.02 ± 0.16 in BRAO. Final VA was 2.12 ± 1.24 in CRAO and could not be calculated in BRAO. The relationship between COVID-19 or its vaccination and RAOs was investigated through past case reports. Two types of reports existed regarding RAO incidence after the COVID-19 pandemic-some indicated an increase, while others found no change. No reports suggested a decrease in RAO occurrence. The current evidence does not clarify the relationship between COVID-19 or its vaccine and RAOs. However, this relationship cannot be ruled out. Further investigations are necessary, as future infectious disease pandemics may occur.

Background The aim of this study was to compare the operation time and patient discomfort between a three‐dimensional heads‐up display system (HUDS) and a conventional ophthalmic surgical microscope (OSM) used in bilateral cataract surgery. Methods This retrospective study included 106 eyes from 53 patients who underwent immediate sequential bilateral cataract surgery at a single tertiary referral center. HUDS and OSM were used alternately for each eye of the same patient. The total operation and procedure times were measured using video recordings. Complications, intraocular pressure, corneal edema, and pain‐related subjective discomfort were assessed using electronic medical records. Best‐corrected visual acuity and prediction error were evaluated 1 month after surgery. Results There were no significant differences in the cataract grade or biometric parameters between the groups. The HUDS and OSM groups showed comparable results in terms of the total operation time, procedure time, cumulative dissipated energy, and irrigation volume. Posterior capsule rupture occurred in one patient (1.89%) in the HUDS group. On the first postoperative day, there were no significant differences in intraocular pressure, corneal edema, or patient‐reported discomfort between the groups. Best‐corrected visual acuity and prediction error were also comparable at 1 month postoperatively. Conclusions In this bilateral comparison study, HUDS and OSM demonstrated similar efficacy and patient‐reported discomfort, with comparable early and 1‐month postoperative outcomes.

The increasing prevalence of vision-threatening retinal diseases poses a growing challenge on healthcare systems worldwide, underscoring the need for effective strategies in disease diagnosis and management. Unsupervised artificial intelligence (AI) approaches, such as anomaly detection, offer the potential to identify established and novel retinal imaging biomarkers. They reduce reliance on expert-driven annotations, mitigate bias, and overcome the limitations of predefined disease categories. Moreover, when integrated into clinical workflows, they have the potential to enhance practical applicability by supporting efficient screening, longitudinal disease monitoring, and objective assessment of treatment response. This review examines the role of anomaly detection for discovering clinically relevant retinal imaging biomarkers, by categorizing biomarkers according to their underlying pathophysiology, summarizing the spectrum of anomaly detection methods applied in the field, and highlighting biomarkers that have been identified through these approaches. Current anomaly detection methods favor the identification of established biomarkers characterized by strong intensity contrasts and well-defined structural boundaries, while more subtle abnormalities remain difficult to capture. Future research should prioritize integrating large language models (LLMs), foundation methods, multimodality, and few-shot anomaly detection, and increase explainability to advance the clinical application of anomaly detection for retinal imaging biomarker discovery in ophthalmology.

This study aimed to evaluate changes in systemic inflammatory indices and retinal structural parameters in newly diagnosed diabetic retinopathy (DR) patients before and after intravitreal anti-VEGF treatment. A total of 308 participants (157 DR patients and 151 healthy controls) were enrolled. Systemic inflammatory status was assessed using hematologic indices derived from routine complete blood count parameters. Optical coherence tomography (OCT) was used to measure macular thickness. DR patients were evaluated at three time points: before treatment, 1 month, and 3 months after anti-VEGF injection. Baseline inflammatory indices (NLR, MLR, PLR, and SII) and macular thickness were significantly higher in the DR group compared to controls (p < 0.001). Following treatment, a progressive decrease in both inflammatory indices and macular thickness was observed. Pretreatment NLR and SII values showed high diagnostic performance in differentiating DR (AUC: 0.925 and 0.838, respectively). Multivariate analysis revealed that NLR, MLR, and HbA1c were independent predictors of macular thickness. Systemic inflammatory activity is closely associated with both the onset and progression of DR. The evaluation of inflammatory indices such as NLR and SII together with OCT findings may support diagnosis and treatment follow-up in clinical practice.

PurposeThis study evaluated the impact of clear corneal incision (CCI) morphology on corneal endothelial cell density (ECD), astigmatism, and higher‐order aberrations (HOAs) following implantable collamer lens (ICL) implantation.DesignProspective study.MethodsSixty‐five patients (65 eyes) undergoing ICL implantation were included. CCI characteristics were assessed with the AS‐OCT CASIA 2, and HOAs were measured using the Pentacam. Corneal astigmatism changes were analyzed using Aplin’s vector analysis.ResultsThe study divided the eyes into two groups: intact incision (29 eyes, with well‐aligned wound architecture) and defective incision (36 eyes, with misaligned incision planes or endothelial gaps). At 3 months postoperatively, corneal thickness at the incision exit (CT‐Ex) was significantly different between the groups (p = 0.038). The incision exit angle (Angle‐Ex) also differed significantly at 1 month (p = 0.023) and 3 months (p = 0.022). The defective incision group showed increases in incision length (IL) and the distance from the entry incision to the center (Dis‐En) between 1 and 3 months (p = 0.006 and p = 0.034, respectively). Both groups showed significant reductions in CT‐Ex, CT‐En, and Angle‐Ex at 3 months. In the defective incision group, trefoil, quadrafoil, and spherical aberrations increased significantly (p < 0.05) while ECD decreased significantly (p < 0.05).ConclusionsShorter IL and smaller exit angle during ICL surgery increase the risk of forming defective incisions, leading to more HOAs and greater ECD loss within the 3‐month postoperative period.

The diagnosis of ocular syphilis in patients with acquired immunodeficiency syndrome (AIDS) remains challenging. This study aimed to evaluate the potential utility of aqueous humor (AH) RPR testing as an adjunctive diagnostic tool for assessing clinical manifestations and visual function in AIDS-associated ocular syphilis. We conducted a retrospective observational cohort study of 31 patients (42 eyes) with ocular syphilis and AIDS treated at Beijing You'an Hospital between 2022 and 2025, based on comprehensive ophthalmic and systemic evaluations. In patients with concurrent syphilis and AIDS, a higher pretreatment AH RPR reactivity level was associated with higher vitreous haze grades (OR = 4.10, 95% CI: 1.62-10.38, p = 0.003). Pretreatment serum, CSF, and AH RPR titers individually predicted macular chorioretinal involvement (AUC 0.76-0.85) and optic nerve involvement (AUC 0.62-0.78). Combining all three markers improved prediction for macular chorioretinal involvement (AUC 0.88) and for optic nerve involvement (AUC 0.79). RPR titers in serum, CSF, and AH were all significantly correlated with both pre- and posttreatment best-corrected visual acuity, indicating that higher titers were associated with worse visual function. AH RPR can serve as a valuable adjunct for diagnosing ocular syphilis, as it indicates intraocular disease activity and informs treatment recommendations for the condition. Trial Registration: Chinese Clinical Trial Registry: 2200056954.

The rapid spread of unreliable misinformation on vision-threatening diseases can significantly affect the eye health behaviors and outcomes of the patients consuming short-form social media content. This study evaluates short-form videos pertaining to vision-threatening diseases to quantify video quality, content, and popularity. This cross-sectional study analyzed short-form videos on cataracts, diabetic retinopathy, glaucoma, and age-related macular degeneration from TikTok, Instagram Reels, and YouTube Shorts. A hashtag search identified the first fifty videos on each disease from each social media platform. Two reviewers evaluated them, resolving discrepancies with a third. Outcome measures included number of views, likes, comments, uploader source, content type, modified DISCERN score (0-5 scale), and global quality scale (GQS) score (1-5 scale). Engagement outcomes were summarized descriptively using medians and interquartile ranges, while reliability and quality outcomes were analyzed using one-way ANOVA with Tukey post hoc comparisons. TikTok videos demonstrated higher median engagement (views, likes, and comments) compared to Instagram Reels and YouTube Shorts. Videos on cataracts had higher engagement statistics compared to the other vision-threatening diseases across all platforms. Physicians were the most common video source (45%). The most common content categories were treatments/management (36%) and general symptoms (22%). YouTube Shorts had a significantly greater average DISCERN (2.93 ± 0.70) and GQS score (3.85 ± 1.26) than Instagram Reels and TikTok (p < 0.001). Videos from patients had the lowest mean DISCERN and GQS scores. TikTok had the greatest median engagement levels, while YouTube Shorts had the greatest mean quality and reliability. Videos from patients and philanthropists had lower quality scores, while healthcare professionals and organizations had the highest. Future efforts should understand the patients' perspectives, address misinformation, and improve quality across all social media platforms.

PurposeTo compare clinical and corneal epithelial changes between preservative and preservative‐free artificial tear therapy using the MS‐39 in patients with dry eye disease (DED).MethodsThis prospective interventional comparative study included 88 eyes of 44 patients. Preserved artificial tears were given for the left eye (group A) and preservative‐free artificial tears for the right eye (group B). They were evaluated before and 3 months after treatment using MS‐39. The dry eye parameters and corneal epithelial thickness were recorded.ResultsThere was an increase in the NIBUT and improvement in the severity by OSDI score (p-value < 0.001) in each group. There was no significant difference in the mean NIBUT between the two groups (p-value = 0.470). Improvement in the OSDI score (p-value = 0.026) and mean post‐treatment epithelial thickness in the central, paracentral superior, and peripheral inferior epithelial areas were significant in the preservative‐free group than the preservative one (p-value = 0.033, 0.034, and 0.023, respectively).ConclusionAlthough both preservative‐free and preserved artificial tears show an increase in epithelial thickness using MS‐39 and improvement in OSDI, preservative‐free artificial tears show superiority, subjectively and objectively, compared to preserved artificial tears using even lighter preservatives, being a safer option for long‐term application.

Purpose To compare intraocular lens (IOL) power calculation using biometric data from Pentacam AXL vs. IOLMaster 700 in postphotorefractive keratectomy (PRK) patients. Methods Patients with previous PRK and 1 year of postoperative follow‐up were included in the study. Anterior segment parameters were assessed by the Pentacam AXL and IOLMaster 700 devices, and measurements were compared using the paired t‐test. Predicted IOL power was calculated using the Barrett True‐K no history, Shammas‐PL, Haigis‐L, and average ASCRS formulas. The interchangeability of the devices was assessed by calculating the 95% limits of agreement (95% LoA) and the concordance correlation coefficient (CCC). Results A total of 39 eyes of 39 patients were included. The mean difference (MD; Pentacam–IOLMaster) and CCC for mean K, AL, ACD, CCT, and WTW distance were (−0.26 ± 0.19 D [ p &lt; 0.05]; 0.983), (0.01 ± 0.06 mm; 0.998), (0.009 ± 0.035 mm; 0.993), (1.5 ± 7.2 μm; 0.983) and (−0.27 ± 0.13 mm [ p &lt; 0.05]; 0.705), respectively. The MD, CCC, and 95% LoA for IOL power calculation with the Barrett true‐K no history, Shammas‐PL, Haigis‐L, and average formula were (0.45 ± 0.23 D; 0.933; 0.01–0.90), (‐0.25 ± 0.48 D; 0.918; −1.20 to 0.70), (0.33 ± 0.27 D; 0.952; −0.20–0.87) and (0.18 ± 0.27 D; 0.970; −0.35–0.72), respectively. All p values for MD of IOL power calculation were less than 0.05. Conclusions In post‐PRK eyes, Pentacam and IOLMaster showed substantial agreement for AL, ACD, and CCT but not for Km and WTW distance, with the latter showing very poor agreement. The two devices could not be used interchangeably for IOL power calculation with any of the 4 studied formulas; however, the strongest to the weakest agreement was observed for the Average formula, Haigis‐L, Barrett, and Shammas‐PL formulas, respectively.

Eye injury is a leading cause of blindness and vision loss worldwide, with children and adolescents being particularly vulnerable. This study aimed to evaluate the global, regional, and national burden of eye injuries among children and adolescents from 1990 to 2021. This is a population-based analysis using data from Global Burden of Disease Study 2021. Eye injury estimates for children and adolescents aged under 20 years, including incidence and years lived with disability (YLDs), were collected and then analyzed by age, sex, and location. The association between sociodemographic index (SDI) and eye injury burden was also investigated. From 1990 to 2021, the age-standardized incidence rate (ASIR) and age-standardized YLD rate (ASYR) for child and adolescent eye injuries decreased by 21.6% and 21.4%, respectively. In 2021, an estimated 11,547,996 children and adolescents worldwide suffered from eye injuries, causing 84,790 YLDs. Boys and adolescents aged 15-19 years had a higher burden. Regionally, South Asia had the highest number of incident cases in 2021, whereas Australasia reported the highest ASIR and ASYR. Nationally, India had the most incident cases, whereas New Zealand had the highest ASIR and ASYR. The SDI-based analysis revealed that children and adolescents in higher SDI countries had higher ASIR and ASYR than their counterparts in lower SDI countries. Unintentional injuries were the leading cause of child and adolescent eye injuries globally. Despite a decline in age-standardized rates of eye injuries globally, eye injury burden remained substantial for children and adolescents. The burden varied by sex, age, location, and SDI, which highlighted the importance of targeted prevention strategies.