Current Eye Research最新文献

Purpose: To examine the performance of deep-learning models that predicts the visual acuity after cataract surgery using preoperative clinical information and color fundus photography (CFP).

Methods: We retrospectively collected the age, sex, and logMAR preoperative best corrected visual acuity (preoperative-BCVA) and CFP from patients who underwent cataract surgeries from 2020 to 2021 at National Taiwan University Hospital. Feature extraction of CFP was performed using a pre-existing image classification model, Xception. The CFP-extracted features and pre-operative clinical information were then fed to a downstream neural network for final prediction. We assessed the model performance by calculating the mean absolute error (MAE) between the predicted and the true logMAR of postoperative BCVA. A nested 10-fold cross-validation was performed for model validation.

Results: A total of 673 fundus images from 446 patients were collected. The mean preoperative BCVA and postoperative BCVA was 0.60 ± 0.39 and 0.14 ± 0.18, respectively. The model using age and sex as predictors achieved an MAE of 0.121 ± 0.016 in postoperative BCVA prediction. The inclusion of CFP as additional predictor in the model (predictors: age, sex and CFP) did not further improve the predictive performance (MAE = 0.120 ± 0.023, p = 0.375), while adding the preoperative BCVA as an additional predictor resulted in a 4.13% improvement (predictors: age, sex and preoperative BCVA, MAE = 0.116 ± 0.016, p = 0.013).

Conclusions: Our multimodal models including both CFP and clinical information achieved excellent accuracy in predicting BCVA after cataract surgery, while the learning models input with only clinical information performed similarly. Future studies are needed to clarify the effects of multimodal input on this task.

Purpose: Fungal keratitis (FK) is a difficult condition to treat, particularly in underdeveloped nations. The study aimed to compare the in vitro activity of chlorhexidine (CHX) and seven antifungal agents against a collection of fungi recovered from FK patients.

Methods: Seventy-three fungi were collected from patients with FK included in study. The antifungal agents tested were fluconazole, voriconazole, posaconazole, miconazole, natamycin, amphotericin B, and caspofungin. The concentration range for CHX was 1-1024 μg/mL. Assessments of antifungal susceptibility were conducted using the EUCAST broth microdilution reference method.

Results: The findings demonstrated the beneficial in vitro inhibition of filamentous and yeast fungi by CHX. CHX demonstrated efficacy against Fusarium spp., Aspergillus spp., Candida spp., S. apiospermum and dematiceous fungi at concentrations of 4-64, 32-64, 4-32, 8, and 4-16 µg/mL respectively. The median MICs and MIC distributions of CHX showed no significant differences among the evaluated spp. (p > 0.05). The most effective antifungal drug was posaconazole, which was followed by voriconazole, caspofungin, and amphotericin B.

Conclusion: In situations where access to a range of antifungal medications and microbiological facilities is limited, CHX should be further investigated as a potential treatment for FK. It might be able to treat the condition as an inexpensive topical treatment.

Purpose: Diabetic retinopathy (DR) is the most common complication of diabetes mellitus. Stimulator of interferon genes (STING) plays an important regulatory role in the transcription of several genes. This study aimed to mine and identify hub genes relevant to STING in DR.

Methods: The STING-related genes (STING-RGs) were extracted from MSigDB database. Differentially expressed STING-RGs (DE-STING-RGs) were filtered by overlapping differentially expressed genes (DEGs) between DR and NC specimens and STING-RGs. A PPI network was established to mine hub genes. The ability of the hub genes to differentiate between DR and NC specimens was evaluated. Additionally, a ceRNA network was established to investigate the regulatory mechanisms of hub genes. Subsequently, the discrepancies in immune infiltration between DR and NC specimens were further explored. Additionally, we performed drug predictions. Finally, RT-qPCR of peripheral blood samples was used to validate the bioinformatics results.

Results: A grand total of four genes (IKBKG, STAT6, NFKBIA, and FCGR2A) related to STING were identified for DR. The AUC values of all four hub genes were greater than 0.7, which indicated that the diagnostic value was acceptable. The ceRNA network contained four hub genes, 170 miRNAs, and 135 lncRNAs. In addition, immunoinfiltration analysis demonstrated that the abundance of activated B cells was notably different between the DR and NC specimens. Moreover, 32 drugs were included in the drug-gene network, with twelve drugs targeting STAT6, nine drugs targeting NFKBIA, four drugs targeted IKBKG, and seven drugs targeted FCGR2A. The expression of the four hub genes in blood samples determined by RT-qPCR was consistent with our analysis.

Conclusion: In conclusion, four hub genes (IKBKG, STAT6, NFKBIA, and FCGR2A) related to STING with a diagnostic value for DR were identified by bioinformatics analysis, which might provide new insights into the evaluation and treatment of DR.

Purpose: To identify risk factors and effect modifiers associated with intraocular inflammation (IOI) following brolucizumab injection.

Methods: Our protocol was registered on PROSPERO (CRD42022382645). We searched six electronic databases (PubMed, Scopus, Web of Science, CENTRAL, EMBASE, and Google Scholar) to retrieve all studies that reported the occurrence of IOI following brolucizumab. Data are reported as mean difference with their corresponding 95% confidence intervals. All analyses were conducted per eye, and the risk of bias was assessed using the National Health Institute tool.

Results: Our analysis included 3527 eyes of 3469 patients of 33 papers. The mean age of the patients was 74 years (SD = 10.9, Range = 62.3-80.9). There were 1793 male patients (51.7%) and 1719 female patients (49.6%). The average follow-up period was 13.9 months (SD = 9.4). The mean number of injections was 4.5 (SD = 2.9) injections per eye; 1315 (37.3%) eyes had neovascular AMD, 189 (5.4%) had diabetic macular edema, and 129 (3.7%) eyes had polypoidal choroidal vasculopathy. Post-intervention, subretinal fluid, intraretinal fluid, and pigment epithelial detachment were significantly improved (46.5-11.3% of patients, 55.7-11.3% of patients, 24.7-7.1% of patients, respectively) (p < 0.001). Regarding visual acuity, there was an improvement with a mean difference of 0.12 (95% CI = 0.18-0.07, z = 4.38, p < 0.0001, 2064 eyes). The most common reported complication is IOI (n = 196, 6%). IOI was observed more in the elderly (76.3 ± 9.2 years), females (66%), and after the second injection.

Conclusions: This systematic review provides valuable insights into risk factors and effect modifiers for IOI associated with brolucizumab treatment, aiding clinicians in optimizing patient care. Future studies should prioritize prospective, long-term investigations to further elucidate the safety profile of brolucizumab and refine its use in the management of retinal and choroidal vascular diseases.

Purpose: To investigate the longitudinal relationship between myopia and retinal shape in Chinese children.

Methods: A total of 2471 seven-year-old Chinese children were measured for axial length, anterior corneal radius of curvature, anterior chamber depth, lens thickness, central spherical equivalent, and peripheral refractions along the horizontal meridian (±15°, ±30°) under cycloplegia. Retinal shape was fitted using vertex radius of curvature, asphericity, and an offset. The areas under the horizontal retinal curve, the nasal side of areas under the horizontal retinal curve and the temporal side of areas under the horizontal retinal curve, were calculated. Children were tested annually for 5 years from year 0 to 4, with 1123 newly developed myopes divided into eight subgroups based on relative time to myopia onset.

Results: In follow-up subgroups, vertex radius of curvature showed positive correlations with central spherical equivalent from years 1 to 4. Smaller temporal side of areas under the horizontal retinal curve was associated with negative central spherical equivalent each year. Myopic shift (Δcentral spherical equivalent) was negatively correlated with baseline central spherical equivalent and with greater change in areas under the horizontal retinal curve from year 0 to 4. The absolute change in temporal side of areas under the horizontal retinal curve (0.39 ± 1.15 mm²) was significantly larger than the absolute change in nasal side of areas under the horizontal retinal curve (0.05 ± 1.11 mm²) over the 4 years. In myopia onset subgroups, temporal side of areas under the horizontal retinal curve was significantly smaller than nasal side of areas under the horizontal retinal curve. The ratio of temporal side of areas under the horizontal retinal curve/nasal side of areas under the horizontal retinal curve was closest to 1 in the year before myopia onset and decreased as myopia developed. Vertex radius of curvature, areas under the horizontal retinal curve, and temporal side of areas under the horizontal retinal curve decreased linearly as myopia progressed.

Conclusion: A progressive steepening of the horizontal posterior retina was associated with myopia progression. Myopia shift was negatively correlated with baseline central spherical equivalent and with greater change in posterior retinal shape. As myopia progressed, the horizontal retina shape displayed increased asymmetry.

Purpose: Astragalus polysaccharide (APS), a water-soluble heteropolysaccharide, possesses immunomodulatory, anti-inflammatory, and cardioprotective properties. This study investigates the neuroprotective potential of APS in a model of N-Methyl-d-aspartic acid (NMDA)-induced retinal neurodegeneration, aiming to explore its potential as a treatment for retinal degenerative diseases.

Methods: Retinal function was evaluated using electroretinography (ERG), optomotor reflex (OMR), and flash visual evoked potentials (FVEP). Retinal inflammatory responses were examined through immunohistochemistry, western blotting (WB), and quantitative reverse transcription PCR (qRT-PCR). To assess the integrity of visual projections, an intravitreal injection of adeno-associated virus (AAV) was employed to trace the projections of retinal ganglion cells (RGCs) to the visual centers.

Results: APS treatment conferred protection to retinal cells, as indicated by ERG and OMR assessments. And APS intervention mitigated NMDA-induced apoptosis, evidenced by a decrease in TUNEL-positive cells. Furthermore, APS treatment attenuated the NMDA-induced reduction in RGC projections to the visual centers, including the superior colliculus and lateral geniculate nucleus, as demonstrated by AAV tracing.

Conclusions: Our findings reveal that APS shields the retina from NMDA-induced damage by inhibiting the NF-κB signaling pathway and reduces the detrimental effects of NMDA on RGC projections to the visual centers. These findings propose APS as a potential novel therapeutic agent for the treatment of retinal diseases.

Purpose: Age-related macular degeneration (AMD) is a chronic retinal disease that can lead to blindness. While the NLR family pyrin domain containing 3 (NLRP3) inflammasome is implicated in AMD, the specific roles of miR-21 and NLRP3 in AMD-related inflammation remain unclear. Therefore, this study aimed to investigate the roles of miR-21 and NLRP3 in blue light-induced neurodegeneration in the mouse retina.

Methods: A mouse model of retinal light damage was established through three months of blue light exposure (BLE). The experimental groups comprised the Control (Ctrl), BLE, BLE + miR-nc, and BLE + miR-21 inhibitor groups. The microRNAs were administered via intravitreal injections once per week. After successful modeling, changes in visual function and retinal morphology were investigated by using electroretinography and hematoxylin and eosin staining, respectively. Photoreceptor apoptosis was assessed using the TdT-mediated dUTP nick-end labeling assay. Immunofluorescence was used to detect and locate microglia and NLRP3 expression in the mouse retina. The expression of miR-21, NLRP3, and downstream factors in the retinas of each group was measured using qRT-PCR and western blotting.

Results: In the BLE and BLE + miR-nc groups, there was a decrease in visual function and retinal thickness, an increase in retinal ganglion cell injury and photoreceptor cell apoptosis, and elevated microglia activity in the retina, as evidenced by their migration to the outer retinal layer. In addition, the expression of miR-21, NLRP3, and downstream factors was increased in the BLE and BLE + miR-nc groups compared to that in the control group. However, intravitreal injection of the miR-21 inhibitor reduced miR-21 expression in the retina and significantly inhibited the activation of the NLRP3 inflammasome, effectively alleviating retinal photodamage caused by BLE.

Conclusions: This study indicates that miR-21 may mitigate blue-light-induced retinal neurodegeneration by reducing the activation of the NLRP3 inflammasome in the mouse retina.

Purpose: This study aimed to explore the expression profile of miR-185-5p in proliferative DR (PDR), and further evaluate its diagnostic value and possible mechanism of miR-185-5p in PDR.

Methods: The level of miR-185-5p was detected by qRT-PCR. The ROC curve was established to estimate the diagnostic ability of miR-185-5p. Transwell experiment and cell counting kit-8 (CCK-8) assays were conducted to assess the effect of miR-185-5p on the migration and proliferation of human retinal endothelial cells (HRECs) induced by high glucose. Enzyme linked immunosorbent assay (ELISA) was used to detect the concentrations of inflammatory factors. The luciferase reporter gene experiment was used to prove the interaction between miR-185-5p and CXCR4.

Results: Compared to the control group, the expression of miR-185-5p was significantly up-regulated in both the type 2 diabetes mellitus (T2DM) group and the PDR groups, with higher levels in the PDR group than in the T2DM group. The ROC curve reveals that serum miR-185-5p can distinguish PDR patients from T2DM patients. MiR-185-5p levels in HRECs increased significantly after high glucose induction. High glucose induction also promoted the migration, proliferation and inflammation response of HRECs. However, when the intracellular miR-185-5p level was down-regulated by miR-185-5p inhibitor transfection, these effects were inhibited. The luciferase reporter gene assay showed that miR-185-5p directly targets CXCR4.

Conclusion: The expression of miR-185-5p is out of balance in PDR and it may be involved in regulating the migration and proliferation of HRECs by regulating CXCR4.

Purpose: With age, the mammalian lens forms successive layers of crystallin protein fibers which infoliate with lens growth and development. As heavy metals generally bind to tissue protein, heavy metals are posited to sequester within the lens with age. Therefore, this study aims to compare heavy metals in human crystalline lens of older adults to known physiologic blood and urine levels and assess the association between concentrations in the lens and metabolic biomarkers.

Methods: Consecutive lens specimens obtained during cataract surgery by phacoemulsification were subjected to atomic spectrometry for heavy metal content. A one-sample t-test compared heavy metals in lens to known physiologic blood and urine concentrations. Linear regression models assessed the association between heavy metals and biomarkers of metabolic function. Linear discriminant analysis assessed the classification of gender and smoking status based on multiple and individual heavy metals.

Results: All heavy metal levels were elevated in lens specimens compared to blood and urine with the exception of iron (p < 0.0001). Lens titanium and copper were positively associated with blood-urea nitrogen (Titanium: $\hat{\beta }$ = 1.14, p = 0.04, Copper: $\hat{\beta }$ = 1.12, p = 0.03. Lens copper was positively associated with creatinine ($\hat{\beta }$ = 1.10; p = 0.02), but negatively associated with glomerular filtration rate ($\hat{\beta }$ = 0.89; p = 0.02). Lens chromium and lead were positively associated with albumin (Chromium: $\hat{\beta }$ = 1.03, p = 0.03; Lead: $\hat{\beta }$ = 1.02, p = 0.04). Lens nickel was positively associated with bilirubin ($\hat{\beta }$ = 1.14; p = 0.03). Classification based on multiple or individual heavy metals for gender and smoking status was not statistically significant.

Conclusions: Our results suggest the human crystalline lens accumulates heavy metals with age and demonstrate the correlation between abnormality of metabolic function and heavy metal deposition in older adult lens.

Purpose: To compare the visual outcomes, rate of cystoid macular edema (CME), and additional associated complications in eyes that exhibited zonular dialysis (ZD) during phacoemulsification to a reference group of uneventful phacoemulsification eyes.

Methods: A retrospective multicenter comparative database study. We pooled data from 8 United Kingdom sites between 2003 and 2015. The main outcome measures were the mean postoperative visual acuity (VA) at 12-24 weeks and the rates of CME and additional associated complications.

Results: We included 1074 eyes in the ZD group and 112,479 in the reference group. Logistic regression analysis showed that pseudoexfoliation was the strongest associated factor of ZD (OR: 6.1), followed by previous glaucoma surgery (OR: 4.4). Mean logMAR preoperative VA was 0.8 ± 0.6 in the ZD group vs. 0.6 ± 0.5 in the reference group (p < 0.001). Mean postoperative VA was worse in the ZD group (p < 0.001); 0.4 ± 0.6 vs. 0.2 ± 0.3 and 0.5 ± 0.6 vs. 0.2 ± 0.3 at 4-12 weeks and 12-24 weeks, respectively. At 12-24 weeks, the proportions of eyes that gained ≥0.3 logMAR units were 50% in the ZD group vs. 62% in the reference group (p < 0.001). In the ZD group, the most common intraoperative complication was vitreous loss (34.3%), followed by posterior capsular rupture (PCR) (11.1%). Postoperative CME occurred in 2.3% vs. 1.4% (p = 0.01), and 9.3% of eyes required surgery for correction of aphakia, intraocular lens decentration, or dropped lens figments removal.

Conclusions: The occurrence of ZD was associated with worse postoperative vision, an increased rate of vitreous loss and PCR, and a higher risk of CME.