Abstract:Glaucoma is an eye disease characterized by pathologically elevated intraocular pressure, optic nerve atrophy, and visual field defects, which can lead to irreversible vision loss. In recent years, the rapid development of artificial intelligence (AI) technology has provided new approaches for the early diagnosis and management of glaucoma. By classifying and annotating glaucoma-related images, AI models can learn and recognize the specific pathological features of glaucoma, thereby achieving automated imaging analysis and classification. Research on glaucoma imaging classification and annotation mainly involves color fundus photography (CFP), optical coherence tomography (OCT), anterior segment optical coherence tomography (AS-OCT), and ultrasound biomicroscopy (UBM) images. CFP is primarily used for the annotation of the optic cup and disc, while OCT is used for measuring and annotating the thickness of the retinal nerve fiber layer, and AS-OCT and UBM focus on the annotation of the anterior chamber angle structure and the measurement of anterior segment structural parameters. To standardize the classification and annotation of glaucoma images, enhance the quality and consistency of annotated data, and promote the clinical application of intelligent ophthalmology, this guideline has been developed. This guideline systematically elaborates on the principles, methods, processes, and quality control requirements for the classification and annotation of glaucoma images, providing standardized guidance for the classification and annotation of glaucoma images.

Abstract:AIM: To predict the post-operative vault and the suitable size of the implantable collamer lens (ICL) by comparing the performance of multiple artificial intelligence (AI) algorithms. METHODS: A retrospective analysis of 83 patients with 132 eyes was conducted from 2020 to 2023. All patients underwent implantation of EVO-V4C ICLs. ICLs were selected based on STAAR’s recommended formula. Postoperative vault values were measured using anterior segment optical coherence tomography (ASOCT). First, feature selection was performed on patients’ preoperative examination parameters to identify those most closely related to postoperative vault and incorporate them into the machine learning model. Subsequently, four regression models, namely MLP, XGBoost, RFR, and KNN, were employed to predict the vault, and their predictive performances were compared. The ICL size was set as the prediction target, with the vault and other input features serving as new inputs for predicting the ICL size. RESULTS: Among all preoperative parameters, 16 parameters were most closely related to postoperative vault and were included in the prediction model. In vault prediction, XGBoost performed the best in the regression model (R²=0.9999), followed by MLP (R²=0.9987) and RFR (R²=0.8982), while the KNN model had the lowest predictive performance (R²=0.3852). XGBoost achieved a prediction accuracy of 99.8%, MLP had a prediction accuracy of 98.9%, while RFR and KNN had accuracies of 87.1% and 57.4%, respectively. CONCLUSION: AI effectively predicts postoperative vault and determines ICL size. XGBoost outperforms other machine-learning algorithms tested. Its accurate predictions help ophthalmologists choose the right ICL size, ensuring proper vaulting.

Abstract:AIM: To assess the performance of five distinct large language models (LLMs; ChatGPT-3.5, ChatGPT-4, PaLM2, Claude 2, and SenseNova) in comparison to two human cohorts (a group of funduscopic disease experts and a group of ophthalmologists) on the specialized subject of funduscopic disease. METHODS: Five distinct LLMs and two distinct human groups independently completed a 100-item funduscopic disease test. The performance of these entities was assessed by comparing their average scores, response stability, and answer confidence, thereby establishing a basis for evaluation. RESULTS: Among all the LLMs, ChatGPT-4 and PaLM2 exhibited the most substantial average correlation. Additionally, ChatGPT-4 achieved the highest average score and demonstrated the utmost confidence during the exam. In comparison to human cohorts, ChatGPT-4 exhibited comparable performance to ophthalmologists, albeit falling short of the expertise demonstrated by funduscopic disease specialists. CONCLUSION: The study provides evidence of the exceptional performance of ChatGPT-4 in the domain of funduscopic disease. With continued enhancements, validated LLMs have the potential to yield unforeseen advantages in enhancing healthcare for both patients and physicians.

Abstract:AIM: To investigate the associations between urinary dialkyl phosphate (DAP) metabolites of organophosphorus pesticides (OPPs) exposure and age-related macular degeneration (AMD) risk. METHODS: Participants were drawn from the National Health and Nutrition Examination Survey (NHANES) between 2005 and 2008. Urinary DAP metabolites were used to construct a machine learning (ML) model for AMD prediction. Several interpretability pipelines, including permutation feature importance (PFI), partial dependence plot (PDP), and SHapley Additive exPlanations (SHAP) analyses were employed to analyze the influence from exposure features to prediction outcomes. RESULTS: A total of 1845 participants were included and 137 were diagnosed with AMD. Receiver operating characteristic curve (ROC) analysis evaluated Random Forests (RF) as the best ML model with its optimal predictive performance among eleven models. PFI and SHAP analyses illustrated that DAP metabolites were of significant contribution weights in AMD risk prediction, higher than most of the socio-demographic covariates. Shapley values and waterfall plots of randomly selected AMD individuals emphasized the predictive capacity of ML with high accuracy and sensitivity in each case. The relationships and interactions visualized by graphical plots and supported by statistical measures demonstrated the indispensable impacts from six DAP metabolites to the prediction of AMD risk. CONCLUSION: Urinary DAP metabolites of OPPs exposure are associated with AMD risk and ML algorithms show the excellent generalizability and differentiability in the course of AMD risk prediction.

Abstract:AIM: To investigate the capabilities of large language models (LLM) for providing information and diagnoses in the field of neuro-ophthalmology by comparing the performances of ChatGPT-3.5 and -4.0, Bard, and Bing. METHODS: Each chatbot was evaluated for four criteria, namely diagnostic success rate for the described case, answer quality, response speed, and critical keywords for diagnosis. The selected topics included optic neuritis, non-arteritic anterior ischemic optic neuropathy, and Leber hereditary optic neuropathy. RESULTS: In terms of diagnostic success rate for the described cases, Bard was unable to provide a diagnosis. The success rates for the described cases increased in the order of Bing, ChatGPT-3.5, and ChatGPT-4.0. Further, ChatGPT-4.0 and -3.5 provided the most satisfactory answer quality for judgment by neuro-ophthalmologists, with their sets of answers resembling the sample set most. Bard was only able to provide ten differential diagnoses in three trials. Bing scored the lowest for the satisfactory standard. A Mann-Whitney test indicated that Bard was significantly faster than ChatGPT-4.0 (Z=-3.576, P=0.000), ChatGPT-3.5 (Z=-3.576, P=0.000) and Bing (Z=-2.517, P=0.011). ChatGPT-3.5 and -4.0 far exceeded the other two interfaces at providing diagnoses and were thus used to find the critical keywords for diagnosis. CONCLUSION: ChatGPT-3.5 and -4.0 are better than Bard and Bing in terms of answer success rate, answer quality, and critical keywords for diagnosis in ophthalmology. This study has broad implications for the field of ophthalmology, providing further evidence that artificial intelligence LLM can aid clinical decision-making through free-text explanations.

Abstract:AIM: To explore the immune cell infiltration and molecular mechanisms of retinal ischemia-reperfusion injury (RIRI) to identify potential therapeutic targets. METHODS: In the bulk RNA-seq analysis, This study performed differential gene expression analysis, weighted gene co-expression network analysis, and protein-protein interaction network analysis to identify hub genes. QuanTIseq was used to determine the composition of infiltrating immune cells. Following the identification of hub genes, single-cell RNA-seq analysis was employed to pinpoint the specific immune cell types expressing these hub genes. Cell-cell communication analysis to explore signaling pathways and interactions between immune cells was further performed. Finally, the expression of these key immune regulators in vivo using quantitative real-time polymerase chain reaction (qRT-PCR) was validated. RESULTS: Bulk RNA-seq analysis identified Stat2, Irf7, Irgm1, Igtp, Parp9, Irgm2, Nlrc5, and Tap1 as hub genes, with strong correlations to immune cell infiltration. Single-cell RNA-seq analysis further revealed six immune cell clusters, showing Irf7 predominantly in microglia and Tap1 in dendritic cells (DCs). And cell-cell communication analysis showed that microglia and DCs play central roles in coordinating immune activity. qRT-PCR validated the upregulation of these genes. CONCLUSION: In the acute phase of RIRI, Irf7 and Tap1 may be the potential therapeutic targets to reduce inflammation and promote neurological function recovery.

Abstract:AIM: To investigate the effects and the underlying mechanism(s) of conbercept on the phagocytosis of hard exudates (HEs) by Müller glia in diabetic retinopathy (DR). METHODS: Twenty-one eyes from 17 patients with diabetic macular edema (DME) underwent optical coherence tomography (OCT) imaging to examine the changes of HEs before and after intravitreal conbercept injection (IVC). In vitro, rat retinal Müller cell line (rMC-1) was cultured under high glucose and treated with oxidized low-density lipoprotein (Ox-LDL) with or without conbercept. Phagocytosis was analysed with immunofluorescence, flow cytometry, and Western blot. Expressions of scavenger receptors (LOX-1, CD36) were analyzed by quantitative real-time polymerase chain reaction (qRT-PCR). Conbercept’s effects on vascular endothelial growth factor A (VEGF-A), VEGFR2, inflammation (NF-κB, IL-6, iNOS), and oxidative stress (ROS) were evaluated with Western blot and immunofluorescence. RESULTS: The area of HEs showed minimal change after the first IVC (1.39±1.41 to 1.38±1.3 mm2, P=0.938), but significantly decreased after the third IVC (0.45±0.66 mm2, P=0.002). In vitro, conbercept enhanced the phagocytosis of Ox-LDL by rMC-1 cells under high glucose condition. Conbercept reduced ROS and inflammation (NF-κB, IL-6, iNOS) in high glucose-treated rMC-1 cells by suppression of VEGF/VEGFR2 pathway. The inhibition of NF-κB by conbercept further activated PPARγ-CD36 axis, increasing CD36 expression and promoting Ox-LDL uptake, thereby facilitating the clearance of HEs. CONCLUSION: Conbercept reduces HEs in DR by enhancing Müller glia phagocytosis possibly through activating PPARγ-CD36 axis, which is mediated by inhibition of VEGF signaling. Modulation of Müller glia phagocytic capacity might provide a novel therapeutic strategy to treat DR and DME.

Abstract:AIM: To analyze the pathogenicity and clinical features of patients in a consanguineous cone-rod dystrophy (CRD) family due to heterozygous variants in the GUCY2D gene. METHODS: Whole exome sequencing was used to screen for pathogenic genes and candidate pathogenic variants were obtained by bioinformatics analysis. Sanger sequencing was used for validation and familial co-segregation analysis to determine pathogenic variants. Pymol software was applied to produce a 3D structure image of the protein to analyze the structural and functional alterations of the protein. The pathogenicity of genetic variants was evaluated according to ACMG guidelines. RESULTS: The chief clinical symptoms of this proband included obvious visual impairment, protanopia and deuteranopia, peripheral punctate pigment, arteriolar attenuation, structural and functional abnormalities revealed by optical coherence tomography (OCT) and electroretinography (ERG) including thinning of the outer retinal layer, a discontinuous external limiting membrane (ELM) and ellipsoid zone (EZ), granular hyperreflective projections between the retinal pigment epithelium and the interdigitation zone, severe attenuation of photopic responses with mild reduced scotopic responses. Whole-exome sequencing revealed that the proband carried a heterozygous variant of the GUCY2D gene: c.2512C>T: p.Arg838Cys. Three-dimensional molecular structure analysis of the protein revealed that amino acid 838 was mutated from polar positively charged arginine to polar uncharged cysteine, and the spatial structure of the protein changed greatly. Sanger sequencing co-segregation analysis confirmed that such a variant was detected in neither the phenotypically normal parents nor the daughter of the proband, which was presumed to be a de novo one. The variant was determined to be pathogenic according to ACMG guidelines. The heterozygous variant at the same site was detected in the abnormal proband’s son with moderate attenuation of photopic electroretinographic responses and normal scotopic electroretinographic responses, supporting autosomal dominant inheritance. CONCLUSION: The de novo variant causing atypical autosomal dominant CRD is identified in a Chinese consanguineous family and this variant passes through this family in an autosomal dominant mode of inheritance, revealing the complex diversity and unpredictability of the inheritance mode for common single-gene genetic disease.

Abstract:AIM: To determine the microbiological agents and antibiotic susceptibility of chronic dacryocystitis in recent years to guide effective treatment strategies. METHODS: A total of 308 adult patients with chronic dacryocystitis were enrolled from January 2020 to September 2022 in Beijing Tongren Hospital. The 229 pus specimens were taken from the conjunctival sac, while 79 specimens were taken from the opened lacrimal sac during transnasal dacryocystorhinostomy (DCR) surgery. All the samples were sent for microbiological smear and culture and drug susceptibility tests. RESULTS: The 202 specimens showed microbial growth, with a positive rate of 65.6%. The 313 strains of bacteria were isolated, including 272 aerobic or facultative anaerobes, accounting for 86.9%, and 41 anaerobic bacteria, accounting for 13.1%. The most common strains were Staphylococcus epidermidis, Corynebacterium macginleyi, and Staphylococcus aureus. Six strains of fungi were isolated. The 14 strains (18.9%) of Staphylococcus were found to be multidrug resistant. It showed a wide variety of gram-negative bacteria, up to 23 species. Specimens obtained during DCR surgery had a positive rate of 70.9%, which was higher than those obtained from the conjunctival sac (63.8%), with a statistically significant difference. More microbiological species were found in intraoperative specimens, and consisted largely of pathogenic bacteria or conditional pathogens. All the 6 strains of fungi were obtained from intraoperative specimens. All bacteria were sensitive to vancomycin, while a large number of bacteria were resistant to erythromycin. The commonly used ocular antibiotics levofloxacin, moxifloxacin and tobramycin were generally sensitive to most of the bacteria. CONCLUSION: Gram-positive cocci, gram-positive bacilli, and gram-negative bacilli are the predominant pathogens for chronic dacryocystitis. Specimens taken during DCR surgery can obtain more accurate microbiological results.

Abstract:AIM: To study the relationship between ultrasound biomicroscopy (UBM) and Lenstar when measuring anterior chamber depth (ACD) in eyes with shallow anterior chamber, and the necessity of conducting both examinations. METHODS: This is a retrospective observational study including 56 acute primary angle closure (APAC) eyes and 47 primary angle closure suspect (PACS) eyes with shallow ACD. ACD value measured by Lenstar and UBM were documented. The Bland-Altman plots were examined separately in all included eyes, APAC eyes and PACS eyes, for the assessment of agreement between two measurements. The agreement was compared across different population by evaluating mean difference, width of 95% limit of agreement (LoA) and the presence of proportional bias or outliers in Bland-Altman plots. RESULTS: The average ACD in APAC eyes (1.71±0.23 mm) was significantly smaller than that in PACS eyes (1.79±0.25 mm, P=0.038). Bland-Altman plots of both APAC eyes and PACS eyes showed small mean difference without the presence of proportional bias. However, compared with PACS eyes, the Bland-Altman plot of APAC eyes had wider 95% LoA and more outliers outside the 95% LoA. CONCLUSION: Despite the small ACD values, the two methods in measuring ACD, Lenstar and UBM, demonstrate substantial consistency when measuring PACS eyes. However, for the APAC population, the agreement might be limited, and thus the measured values should be interpreted with caution.

Abstract:AIM: To investigate the relationship between preoperative corneal biomechanical properties and corneal tomographic properties in cataract patients. METHODS: The study consisted of 59 eyes of 30 participants who were diagnosed as cataract in Peking University Third Hospital between September 2019 and November 2019. Stepwise multivariable linear regression analysis was calculated to determine the relationship between corneal biomechanical parameters and tomographic parameters. The patients were classified into three groups of with the rule (WTR) astigmatism, against the rule astigmatism and oblique astigmatism. And the differences in corneal parameters among different groups were compared. RESULTS: There were significant differences in the first applanation time (A1T), the first applanation length (A1L), corneal velocity during the first applanation (Vin), the second applanation time (A2T), highest concavity (HC) radius, displacement amount (DA), DA ratio, stiffness parameter A1 (SPA1) and integrated radius (IR) between oblique astigmatism patients and the other two groups. Total corneal steep meridian (K2) was negatively associated with A1L, A1T and corneal velocity during the second applanation (Vout). Patients with higher anterior corneal curvature had lower HC radius and central corneal thickness (CCT; P=0.001 and 0.006, respectively), while the Ambrosio relational thickness to the horizontal profile (ARTh) was higher than those with lower anterior corneal curvature (P=0.009). CONCLUSION: The study reveals that the elasticity of corneal collagen fibers is greater, but the viscoelasticity of cornea is smaller in patients with oblique astigmatism. There is no significant difference in ARTh between patients with different types of astigmatism, that is, the corneal biomechanical specificity of oblique astigmatism group is probably not caused by corneal thickness. Moreover, we find patients with higher anterior corneal curvature has lower HC radius and CCT but higher ARTh than those with lower anterior corneal curvature.

Abstract:AIM: To evaluate the accuracy of intraocular lens (IOL) power calculation formulas with/without preoperative aphakic anterior chamber depth (aph-ACD) in pediatric aphakia. METHODS: A total of 102 pediatric patients (150 eyes) undergoing secondary IOL implantation were divided into two groups (in-the-bag or ciliary sulcus). Prediction error was calculated for 9 IOL power calculation formulas, including: 1) not requiring ACD: Hoffer Q, Holladay 1, SRK/T; 2) usable without or with entering ACD: Barrett Universal II (BUII), Emmetropia Verifying Optical (EVO) 2.0, and Ladas Artificial Intelligence Super (Ladas AI); 3) requiring ACD: Haigis, Kane, and Pearl-DGS. Mean prediction error (ME), mean absolute error (MAE), median absolute error (MedAE) and the percentage of eyes within ±0.25, ±0.50, ±0.75, and ±1.00 D were calculated. RESULTS: For the BUII, EVO 2.0, and Ladas AI, with aph-ACD demonstrated a higher MedAE compared to without aph-ACD (BUII: 1.27 vs 1.13 D, EVO 2.0: 1.26 vs 1.06 D, Ladas AI: 1.30 vs 1.10 D; all P<0.05). Formulas requiring ACD (Haigis, Kane, and Pearl-DGS) exhibited larger MedAE than those not requiring aph-ACD (Hoffer Q, Holladay 1, and SRK/T; P<0.05). In the capsular group, the percentage of eyes within ±1.00 D ranged from 44.83% to 74.14%, and it was 19.57% to 32.61% in the sulcus group. CONCLUSION: The introduction of aph-ACD does not improve the accuracy of IOL calculation for pediatric aphakia, regardless of in-the-bag or sulcus IOL secondary implantation. The relationship between aph-ACD and effective lens position in pediatric aphakia warrants further study.

Abstract:AIM: To characterize a Portuguese population of patients with juvenile idiopathic arthritis (JIA) and to evaluate possible associations between clinical factors and ocular involvement. METHODS: Patients diagnosed with JIA in the previous 20y in Hospital Garcia de Orta were included. Data were assessed from Reuma.pt database. Associations between demographic (age and sex), clinical (articular involvement, extra-articular manifestations, biological therapy), laboratory data [anti-nuclear antibodies, anti-cyclic citrullinated peptide (CCP) antibodies, rheumatoid factor, human leukocyte antigen B27 (HLA-B27), C-reactive protein, erythrocyte sedimentation rate] and ocular involvement were assessed. Statistical analysis was performed using Chi-square for categorical variables and Mann-Whitney test for continuous variables. RESULTS: Totally 91 patients were included, 11 (12%) with previous episodes of uveitis. There was a statistically significant preponderance of early age at JIA diagnosis (mean 4.73 vs 9.58y, P=0.008), antinuclear antibodies positivity (P=0.01), and oligoarticular subtype (P=0.04) in the Juvenile idiopathic arthritis-associated uveitis (JIA-U) group. Ocular complications occurred in 36.4% of patients (n=4): cataracts (n=2), band keratopathy (n=1) and posterior synechiae (n=1). The occurrence of complications was corelated with a shorter period between JIA diagnosis and the first JIA-U episode (mean 0.67 vs 4.88y, P=0.012) but not with age at JIA diagnosis or articular involvement. There was erythrocyte sedimentation rate elevation in the 12mo preceding uveitis (mean 40.5 mm/h, range 13-83). CONCLUSION: The occurrence of JIA-U shortly after JIA diagnosis is shown to be a potential risk factor for ocular complications.

Abstract:AIM: To elucidate causal pathways between oxidative biomarkers and age-related macular degeneration (AMD) phenotypes. METHODS: A bidirectional Mendelian randomization (MR) analytical protocol was implemented, which utilized genome-wide association study (GWAS) summary statistics derived from the IEU OpenGWAS repositories. The investigation focused on 11 oxidative stress markers and AMD phenotypes, encompassing both wet and dry subtypes. The MR methodology incorporated inverse-variance weighted (IVW) calculations, MR-Egger statistical regression, weighted median approximation, and weighted mode assessments to estimate causative relationships. Sensitivity evaluations were conducted to verify result robustness and identify potential pleiotropy. RESULTS: Genetically predicted elevated catalase (CAT) concentrations demonstrated significant associations with heightened risks of overall AMD (IVW OR=1.084, 95%CI: 1.021-1.151, P=0.008) and wet AMD phenotype (IVW OR=1.113, 95%CI: 1.047-1.247, P=0.007). Higher genetically predicted albumin concentrations corresponded with reduced AMD risk (IVW OR=0.827, 95%CI: 0.715-0.957, P=0.013) but increased wet AMD risk (IVW OR=1.229, 95%CI: 1.036-1.458, P=0.018). Reverse MR analysis revealed that genetically predicted dry AMD exhibited significant association with reduced albumin levels (IVW OR=0.987, 95%CI: 0.979-0.996, P=0.004), while wet AMD corresponded with decreased total bilirubin (TBIL) and paraoxonase (PON) activity. CONCLUSION: The results offer strong support for a causal link between markers of oxidative stress and the development of AMD, indicating that oxidative processes play a role in driving the disease progression.

Abstract:AIM: To compare the changes in preschool refractive status, ocular biological parameters, and higher-order aberrations in children with retinopathy of prematurity (ROP) after retinal laser photocoagulation or anti-vascular endothelial growth factor (anti-VEGF) treatment and explore their underlying factors. METHODS: This observational study involved 118 eyes of 59 children, aged 3 to 6y, with ROP followed up between March 2023 and October 2024. They were divided into the laser, anti-VEGF, and anti-VEGF+laser groups. The laser group received a single session of laser photocoagulation. The anti-VEGF group received a single anti-VEGF treatment. The anti-VEGF+laser group received a single anti-VEGF treatment after birth followed by supplementary laser treatment within 2wk to 6mo. Ocular biological parameters were measured using IOL Master 700 and Pentacam HR. Right-eye higher-order aberrations were measured using the OPD-Scan III. Best-corrected visual acuities (BCVA), refractive statuses, ocular biological parameters, and higher-order aberrations were assessed and compared. Multiple linear regression analysis was conducted to evaluate the relationships among ocular biological parameters, higher-order aberrations, spherical equivalent (SE), and treatment methods. Pearson’s correlation coefficients were used to assess the relationships between the SE and higher-order aberrations. RESULTS: The laser group had a higher incidence of myopia and a lower SE than the anti-VEGF group. The incidence of astigmatism and cylindrical power were significantly lower for the anti-VEGF than for the laser and anti-VEGF+laser groups (P<0.05). The anterior corneal surface astigmatism was higher for the laser and anti-VEGF+laser groups than for the anti-VEGF group. The anterior corneal surface K2 and lens thickness were higher for the laser and anti-VEGF+laser groups than for the anti-VEGF group. The whole-eye higher-order aberration root mean square (RMS) values for the right eye were significantly higher for the laser and anti-VEGF+laser groups than for the anti-VEGF group. The whole-eye trefoil RMS values for the right eye were also significantly higher for the laser and anti-VEGF+laser groups than for the anti-VEGF group (P<0.05). Laser treatment was significantly associated with SE, anterior corneal surface curvature and astigmatism, lens thickness, whole-eye high-order aberrations, and whole-eye trefoil (all P<0.05). CONCLUSION: Children with ROP who received laser treatment have higher myopia and astigmatism than those who received anti-VEGF treatment. Children treated with laser or supplementary laser treatment have higher anterior corneal astigmatism, anterior corneal curvature, thicker lenses, whole-eye higher-order aberrations, and whole-eye trefoil. The cause of myopia in children with ROP after laser treatment is increased anterior corneal surface curvature and lens thickness.

Abstract:AIM: To investigate the effects of dipeptidyl peptidase-4 inhibitors (DPP4i) and sodium-glucose cotransporter-2 inhibitors (SGLT2i) on diabetic macular edema (DME) and the need for intravitreal injections (IVT) in patients with type 2 diabetes. METHODS: Data were retrospectively collected from the medical records of patients with diabetic retinopathy (DR) taking either DPP4i or SGLT2i as secondary oral hypoglycemic agents in addition to metformin between January 2019 and July 2022. We compared the prevalence of DME and the need for IVT among patients treated with DPP4i or SGLT2i. Propensity score matching was performed using the following variables: age, duration of diabetes, blood glucose control (HbA1c) level, and severity of DR. RESULTS: A total of 268 patients with DR were included in this study. More DPP4i users needed IVT than SGLT2i users (35.3% vs 18.0%, P=0.011), while the prevalence of DME was not different. The use of SGLT2i was associated with a lower need for IVT than DPP4i [odds ratio (OR) 0.404, 95% confidence interval (CI) 0.198–0.823], and similar trends were observed after propensity score matching (OR 0.419, 95%CI 0.181–0.970). However, this tendency was not significant in multiple logistic regressions. For DME, the use of DPP4i was not a significant risk factor compared to SGLT2i. CONCLUSION: The use of SGLT2i may be associated with a lower need for IVT for overall DR complications, while other factors may contribute to this effect. The effect of SGLT2i on the prevention of DME is not evident.

Abstract:AIM: To elucidate the manifestations and associated complications observed in patients with intraocular silicone oil (SO) emulsification through multiple imaging modalities. METHODS: This single-center, observational, retrospective study included 116 patients who underwent pars plana vitrectomy (PPV) with SO injection for retinal detachment (RD), followed by subsequent SO removal at the Second Hospital of Hebei Medical University from January 2013 to January 2023. Comprehensive records of ophthalmic examinations utilizing multiple imaging techniques were maintained. RESULTS: The study comprised 56 females and 60 males, with a mean age of 52.75±13.6y. The mean follow-up duration for SO tamponade was 9.04±11.33mo (range: 1-84mo). Among the participants, 59 patients were diagnosed with SO emulsification, while 57 patients were in the SO unemulsified group. Patients with SO emulsification had a significantly prolonged SO tamponade duration (P<0.01). Multiple imaging techniques revealed notable signs of SO emulsification and its complications, such as 4 cases (3.4%) with posterior corneal SO-like keratic precipitates (KP) observed by anterior segment photography, 23 cases (19.8%) exhibiting spherical high-reflection signals in the inter-retina, retinal pigment epithelium, or choroid detected by Spectralis spectral domain optical coherence tomography (SD-OCT), 4 cases (3.4%) showing slow movement of emulsified SO droplets within retinal vessels during fluorescein angiography (FFA), plain and enhanced head magnetic resonance imaging (MRI) images of these four patients did not detect emulsified SO in the lateral ventricles, suprasellar cistern, subarachnoid space, third ventricle, fourth ventricle, or other intracranial locations. CONCLUSION: Intraocular emulsified SO can lead to damage in both anterior and posterior segment tissues, encompassing corneal degeneration, cataracts, glaucoma, retinal and choroid inflammation. Objective multiple imaging techniques such as anterior segment photography, SD-OCT, FFA, and MRI offer comprehensive evaluation and diagnosis of SO emulsification and its associated complications.

Abstract:AIM: To characterize the demographic and clinical features of ethambutol optic neuropathy (EON) in an Indonesian patient population and explore prognostic factors for visual recovery. METHODS: Retrospective study of 58 EON patients at an Indonesian eye center (2017-2022). Demographics, ethambutol treatment, ophthalmologic findings were collected. Visual outcomes after ethambutol cessation assessed at 3-12mo. Patients categorized as having visual improvement (≥2 Snellen lines and/or >5° visual field gain) or no improvement. RESULTS: Mean age was 55.5±12.9y, with 56.9% females. Median duration of ethambutol use was 9mo (range: 2-20) at a mean daily dose of 19.7±5.3 mg/kg. At presentation, mean visual acuity was 1.3 logMAR (range: 0-2.5), with normal fundus appearance in 72.4% of eyes. The most common visual field defect was generalized depression (52.4%). After ethambutol cessation, 56.9% of patients had visual improvement. Younger age (50.24±13.8y vs 62.14±8.9y, P<0.01), lower ethambutol dose (17.3±5 vs 23.3±2.9 mg/kg‧d, P<0.01), shorter treatment duration (6.79±2.6mo vs 10.27±1.2mo, P<0.01), and absence of hypertension (16% vs 83%, P=0.012) or kidney disease (0 vs 83%, P<0.01) were associated with higher likelihood of visual improvement. Diabetes did not differ between groups (P=0.889). CONCLUSION: Over half experience visual recovery after ethambutol cessation. Younger age, lower cumulative dose, absence of hypertension or kidney disease predict better visual outcomes.

Abstract:AIM: To compare the visual outcomes and corneal higher-order aberrations (HOAs) of patients with high or low myopic astigmatism after small incision lenticule extraction (SMILE). METHODS: A total of 157 eyes of 157 patients who underwent SMILE were included in this retrospective, nonrandomized, comparative study. All the eyes which were with the rule astigmatism were divided into high astigmatism group (HAG; astigmatism ≤-2.00 D, 73 eyes) and low astigmatism group (LAG; astigmatism ≥-1.00 D, 84 eyes). Visual and refractive examinations were performed, HOAs of the anterior surface, posterior surface, and total cornea of the eyes were evaluated preoperatively and 6mo postoperatively. RESULTS: At the postoperative 6-month follow-up, uncorrected distance visual acuity of 20/20 or better was achieved in 97% and 100% eyes in HAG and LAG respectively and 74% and 100% eyes were within -0.50 D. Vector analysis revealed no significant differences in the correction index (P=0.066), angle of error (P=0.091) or flattening index (P=0.987) between two groups. The magnitude of error was -0.37±0.31 D in HAG and -0.04±0.19 D in LAG (P<0.001). Index of success (IOS) was 0.22±0.09 in the HAG and 0.50±0.46 in the LAG (P<0.001). HOAs of most anterior, posterior and total cornea significantly increased after SMILE, especially the spherical aberration and coma. For HAG, the SMILE procedure induced significantly higher anterior, posterior and total cornea horizontal coma and total corneal total HOAs compared with LAG (P<0.001) and these surgically induced HOAs predominantly originated from the anterior surface of the cornea. CONCLUSION: SMILE surgery induces more HOAs and a mild under-correction of astigmatism in eyes with high astigmatism. The increment in HOAs after SMILE is related to preoperative astigmatism.

Abstract:AIM: To investigate the clinicopathological features of cranial-nasal-orbital communicating lesions and identify key diagnostic indicators for differentiating benign and malignant neoplasms. METHODS: The retrospective cohort study analyzed 74 histologically confirmed cases stratified by anatomical involvement at the Wuhan Union Hospital between January 2010 and December 2020: Group A (orbital-nasal group, n=29), Group B (orbital-cranial group, n=27), and Group C (cranial-nasal-orbital group, n=18). Clinicopathological profiles including symptom presentation, histopathology, and invasion patterns were systematically evaluated. RESULTS: The cohort comprised 49 (66.2%) benign and 25 (33.8%) malignant lesions. Compared with benign lesions, malignant lesions had a shorter onset time (12mo vs 2.5mo, P=0.004) and resulted in poorer vision (0.6 vs 1.53, P=0.025). Headache was reported in 28.6% of patients with benign lesions, but none in those with malignant lesions (P=0.002). Conjunctival congestion and edema were observed in 32.7% of patients with benign lesions and 60% of patients with malignant lesions (P=0.028). The ethmoid sinus was the most frequently invaded site (35 cases). Malignant lesions showed greater invasion in the nasal cavity (28.0% vs 0, P=0.000) and anterior cranial fossa (40.0% vs 8.2%, P=0.003) than benign lesions. The orbital-cranial group was more likely to invade through osseous foramina compared with the orbital-nasal group (P=0.002). Neurogenic tumors predominated benign cases (34.7%), whereas blood derived (28%) and glandular tumors (28%) were most prevalent in malignant subgroups. The proportion of malignant tumors in multi-disciplinary combined surgery was higher than that of benign lesions (61.5% vs 38.5%). CONCLUSION: Malignant cranial-nasal-orbital communicating lesions exhibit distinct clinicopathological signatures characterized by rapid progression, aggressive anterior fossa and nasal region, and severe visual morbidity.

Abstract:AIM: To describe the demographics, clinical characteristics and treatment outcomes of childhood amblyopia in a tertiary eye center in western India. METHODS: This was a retrospective longitudinal hospital-based study of 1382 children aged ≤12y included in the National Institute of OphthalMology AmBlyopia StUdy in Indian Paediatric EyeS (NIMBUS) Study. Data on patient demographics, treatment approach, and best-corrected visual acuity (BCVA) changes were reviewed. RESULTS: The mean age of the study cohort was 4.54±2.46y, with males constituting the majority (55.4%). The cause of amblyopia was refractive error in 73.2%, strabismus in 7.3%, and anisometropia in 6.8% of eyes. The majority of therapies comprised glasses (74.4%), followed by occlusion+glasses (10.3%), occlusion alone (7.3%), and surgery+patching+glasses (5.1%). The mean occlusion time was 2.46±1.14h. After a median follow-up of 10.00 (6–85)mo, the mean BCVA significantly improved from 0.85±0.41 to 0.55±0.42 logMAR. Subgroup analysis revealed BCVA gain for all etiologies, including refractive errors (P<0.001), strabismus (P<0.001), cataract (P<0.001), and ptosis (P<0.001). Additionally, eyes with refractive errors showed significantly better BCVA than eyes with cataracts (P<0.001), strabismus (P<0.001) and marginally better BCVA than eyes with ptosis (P<0.05), both at the baseline and final visit. CONCLUSION: Refractive errors are the commonest cause of amblyopia, followed by strabismus and anisometropia. Timely detection, optimal therapy, and periodic follow-up are crucial in bettering visual acuity regardless of the cause.

Abstract:AIM: To summarize and quantitatively evaluate vasculature alteration of foveal zone in systemic lupus erythematosus (SLE) patients by secondary literature analysis. METHODS: A systematic search of PubMed, Embase, Web of Science, Cochrane Library, CBM, CNKI WanFang Data and VIP was conducted. Studies were about retinal vessel density in SLE patients from January 2000 to April 2023 and valid data were extracted. The Joanna Briggs Institute (JBI) critical appraisal checklist was used to evaluate the cross-sectional studies and prospective studies. The measurement data for combined effect size were weighted mean difference (WMD) and 95% confidence interval (CI). The heterogeneity was evaluated by I2 test. The fixed-effect model was adopted when P>0.1 or I2<50%, and random-effect model was adopted in the contrary. Subgroup and sensitivity analysis were utilized to analyze the sources of heterogeneity. The publication bias was evaluated by Egger tests and funnel plots. RESULTS: A total of 14 studies with 445 subjects and 441 healthy controls from 9 countries were enrolled and 11 studies were included in Meta-analysis. The JBI scores of studies were no less than 14 points. The Meta-analysis results indicated that mean parafoveal superficial vessel density (SVD; WMD=-1.22, 95%CI: -1.67, -0.76), mean perifoveal SVD (WMD=-1.42, 95%CI: -1.95, -0.89), mean whole SVD (WMD=-1.66, 95%CI: -2.53, -0.79), mean parafoveal deep vessel density (WMD=-1.67, 95%CI: -2.75, -0.59) and mean whole deep vessel density (WMD= -4.09, 95%CI: -7.67, -0.52) was significantly lower than the control, while mean foveal SVD (WMD=-1.71, 95%CI: -4.65, 1.24), mean foveal avascular zone (FAZ) area (WMD=0.04, 95%CI: -0.01, 0.09) and mean acircularity index (AI; WMD=0.00, 95%CI: -0.02, 0.02) were not different between SLE patients and controls. Subgroup analysis indicated that the heterogeneity in SVD was partially due to the scanning area. Ocellus or binoculus data contributed partially to the heterogeneity in parafoveal deep vessel density and FAZ area. Sensitivity analysis indicated that the results were robust after changing the analysis model except for foveal SVD and FAZ area. There was no bias in included studies except whole SVD. CONCLUSION: Parafoveal superficial and deep vessel density are significantly lower in SLE patients while FAZ area and AI are not different between SLE patients and the control.

Abstract:In recent years, Mendelian randomization (MR) has been increasingly utilized, leveraging genetic variants as instrumental variables. This approach significantly mitigates confounder effects and reverse causation, precisely clarifying the causal links between exposures and outcomes. MR’s unique advantages have made it instrumental in medicine, especially in elucidating glaucoma’s etiology. It facilitates the identification of potential risk factors, laying the groundwork for developing preventative and therapeutic strategies against glaucoma. Recent MR research has delved into diverse potential glaucoma risk factors, including behavioral habits, metabolic profiles, and their causative linkage to the disease. This review encapsulates MR’s analysis in glaucoma etiology, heralding new avenues for understanding underlying mechanisms and establishing causality.

Abstract:

Abstract:

Abstract: