Search

Purpose: Artificial intelligence (AI) is often trained on images without ocular co‑morbidities, limiting its generalizability. This study aims to evaluate the accuracy of a convolutional neural network (CNN) applied to color fundus photos (CFPs) with simulated cataracts (SCs) in detecting diabetic retinopathy (DR). Methods: A database of 3662 CFPs (from Asia Pacific Tele‑Ophthalmology Society (APTOS) 2019) was used. Using transfer learning, a CNN was trained to classify the training images as either DR or non‑DR. The CNN was then applied to classify the testing images after an SC was applied, using varying degrees of Gaussian blur. Results: Accuracy without SC was 97.0%, sensitivity (Sn) 95.7%, specificity (Sp) 98.3%. For mild SC, accuracy was 93.1%, Sn 91.8%, Sp 94.3%. For moderate SC, accuracy was 62.8%, Sn 31.4%, Sp 95.2%. For severe SC, accuracy was 53.5%, Sn 11.8%, Sp 96.5%. Conclusion: SCs significantly impaired AI accuracy. To prepare AI for clinical use, cataracts and other real‑world clinical challenges affecting image quality must be accounted for. [ABSTRACT FROM AUTHOR]

Purpose: Previous literature has investigated opioid prescription trends in ophthalmology at large, however, little has been done looking at differences between subspecialties. We evaluate if significant trends exist among subspecialties in opioid prescribing patterns. This study aims to illuminate potential over-usage of opioids in ophthalmology that could compromise patient quality of life. Methods: Medicare data and "National Plan and Provider Enumeration System (NPPES) Downloadable File" were queried for cases of ophthalmologists with nonsuppressed opioid prescription data from 2014 to 2019. Ophthalmologists with no subspecialty code or missing regional, gender, degree, or graduation information were excluded. Chi-squared analysis, analysis of variance, t-tests, and multivariate logistic regression were utilized. Results: Five thousand one hundred forty-three physician records were included in analysis, 450 of which were by cornea subspecialists. Most cornea cases were male, graduated before 2005, and practiced in the South. All subspecialties had a significantly increased likelihood of making opioid claims and higher prescription rates compared with cornea (P < 0.050) besides glaucoma (P = 0.357). Only oculoplastics had significantly increased likelihood of greater total supply of opioids compared with cornea (odds ratio [OR] = 22.195, 95% confidence interval [CI] = 12.209–40.350, P < 0.001), while pediatrics (OR = 4.036, 95% CI = 1.377–11.831, P = 0.011) and neuro-ophthalmology (OR = 4.158, 95% CI = 1.237–13.975, P = 0.021) in addition to oculoplastics (OR = 64.380, 95% CI = 26.306–157.560, P < 0.001) were predicted to have significantly greater opioid beneficiaries. Males, the South/Midwest, and graduating before 2005, all were generally associated with increased likelihood of greater total opioid claims, supply, beneficiaries, and prescription rate (P < 0.050). Conclusion: Subspecialty, demographic, chronological, and regional trends exist for opioid prescribing patterns in ophthalmology. [ABSTRACT FROM AUTHOR]

Antibody-drug conjugates (ADCs) are a growing class of chemotherapeutic agents for the purpose of treating cancers that often have relapsed or failed first- and second-line treatments. ADCs are composed of extremely potent cytotoxins with a variety of side effects, one of the most significant being ocular toxicity. The available literature describes these toxicities as varying in severity and in incidence, although with disparate methods of evaluation and management. Some of the most common toxicities include microcyst-like epithelial keratopathy and dry eye. We discuss proposed mechanisms of ocular toxicity and describe the reports that mention these toxicities. We focus on ADCs with the most published literature and the most significant effects on ocular tissue. We propose areas for further investigation and possible ideas of future management. We provide a comprehensive look at the reports of ADCs in current literature to better inform clinicians on an expanding drug class. [ABSTRACT FROM AUTHOR]

Purpose. To investigate whether corneal graft survival could be improved by topical or subconjunctival bevacizumab in a murine model of vascularized high-risk corneal transplantation. Methods. Before corneal transplantation, intrastromal sutures were placed for 2 weeks in the corneas of BALB/c mice, inducing intense angiogenesis. Allogeneic corneal transplantation was performed using C57BL/6 donor mice. Topical bevacizumab (2.5%) was delivered 3 times a day for 3 weeks in one treatment group, and 0.02 mL (0.5 mg) bevacizumab was injected subconjunctivally at days 0, 4, 8, and 15 after transplantation in the other treatment group. The control group received no treatment. Grafts were examined twice a week for 8 weeks by slit-lamp microscopy and were photographed once a week by slit-lamp digital camera and scored for opacity. For assessment of corneal neovascularization (NV), a quantitative method was used to measure three primary metrics including neovascular area, vessel caliber, and neovessel invasion area. Results. Both topical and subconjunctival bevacizumab treatment reduced neovascular area and vessel caliber; however, the regression of corneal NV was more profound when treated subconjunctivally. The mean percentage reduction of neovascular area was 55% (P < 0.05) by week 8 in the subconjunctival treatment group and 33% (P = 0.15) in the topical group. Only subconjunctival bevacizumab treatment resulted in significant regression of neovessel invasion area (P < 0.05). All corneal transplants in both the control and the topical groups were rejected by 4 weeks after transplantation. However, in the subconjunctival treatment group, 33% of corneal grafts survived (P < 0.01). Conclusions. Subconjunctival bevacizumab may offer an adjunctive measure to conventional therapies in preventing graft rejection in high-risk corneal transplantation.

Purpose. To determine the effect of azithromycin (AZM) in a murine model of corneal inflammation. Methods. The effect of topical AZM was studied in murine corneal inflammation. Corneal inflammation was induced by thermal cautery in BALB/c mice. Leukocyte infiltration at different time points was analyzed by flow cytometry. At set time points, real-time polymerase chain reaction was performed to quantify the expression of different inflammatory cytokine transcript in the cornea. Corneal samples were analyzed immunohistochemically for the expression of intercellular adhesion molecule-1 (ICAM-1). Corneal neovascularization (CNV) was induced by micropellet (VEGF-A) placement. Mice were then treated topically with either AZM or vehicle. CNV was evaluated morphometrically. Results. Eyes receiving AZM showed a significant decrease in corneal infiltration compared with the vehicle-treated group. AZM also significantly decreased messenger RNA expression levels of interleukin-1β (IL-1β), tumor necrosis factor-α (TNF-α) and ICAM-1 in the cornea. There was no significant difference in CNV between the AZM- and vehicle-treated groups. Conclusions. After an inflammatory insult, topical AZM significantly reduced leukocyte infiltration into the cornea. This was further supported by an associated decrease in expression of IL-1β, TNF-α, and ICAM-1 in the cornea, indicating AZM may have a potential anti-inflammatory effect on corneal inflammation.