Predicting clinical outcome of sulfur mustard induced ocular injury using machine learning model.

The sight-threatening sulfur mustard (SM) induced ocular injury presents specific symptoms in each clinical stage. The acute injury develops in all exposed eyes and may heal or deteriorate into chronic late pathology. Early detection of eyes at risk of developing late pathology may assist in providing unique monitoring and specific treatments only to relevant cases. In this study, we evaluated a machine-learning (ML) model for predicting the development of SM-induced late pathology based on clinical data of the acute phase in the rabbit model. Clinical data from 166 rabbit eyes exposed to SM vapor was used retrospectively. The data included a comprehensive clinical evaluation of the cornea, eyelids and conjunctiva using a semi-quantitative clinical score. A random forest classifier ML model, was trained to predict the development of corneal neovascularization four weeks post-ocular exposure to SM vapor using clinical scores recorded three weeks earlier. The overall accuracy in predicting the clinical outcome of SM-induced ocular injury was 73%. The accuracy in identifying eyes at risk of developing corneal neovascularization and future healed eyes was 75% and 59%, respectively. The most important parameters for accurate prediction were conjunctival secretion and corneal opacity at 1w and corneal erosions at 72 h post-exposure. Predicting the clinical outcome of SM-induced ocular injury based on the acute injury parameters using ML is demonstrated for the first time. Although the prediction accuracy was limited, probably due to the small dataset, it pointed out towards various parameters during the acute injury that are important for predicting SM-induced late pathology and revealing possible pathological mechanisms.

[Methods of diagnosis and treatment of corneal neovascularization]. / Metody diagnostiki i lecheniya neovaskulyarizatsii rogovitsy.

Corneal neovascularization is one of the most common causes of decreased visual acuity and disability for vision loss, increase in the risk of corneal graft rejection, and appearance of opacifications on the cornea. This article reviews literature on etiological factors of the development of corneal neovascularization, as well as modern methods of diagnosis, conservative and surgical treatment of this pathology.

Corneal neovascularisation following deep anterior lamellar keratoplasty for corneal ectasia: incidence, timing and risk factors.

BACKGROUND: The purpose of this study was to evaluate the incidence, timing and risk factors of corneal neovascularisation (NV) after deep anterior lamellar keratoplasty (DALK) for corneal ectasia. METHODS: This study included 616 eyes who underwent DALK between 2012 and 2020 in two tertiary referral centres. In one centre topical corticosteroids were discontinued after complete suture removal 1 year after surgery, whereas in the other they were discontinued 3-4 months after surgery. The presence and severity of corneal NV was ascertained based on slit lamp photographs. Potential risk factors for corneal NV were evaluated using the Cox proportional hazards model. RESULTS: The cumulative incidence of corneal NV was 8.7% at 1 year after surgery and 13.2% at 5 years. Mean time interval from surgery to development of corneal NV was 12.8±16.2 months, with 68.9% of cases occurring before complete suture removal. Early discontinuation of topical steroids, older age and ocular allergy were associated with an increased risk of developing corneal NV (respectively, HR=2.625, HR=1.019, HR=3.726, all p<0.05). CONCLUSIONS: The risk of corneal NV is higher in the first year following DALK. Early discontinuation of topical steroids, ocular allergy and older age are significant predictors of corneal NV.

Diagnostic Accuracy of Monitoring Tests of Fellow Eyes in Patients with Unilateral Neovascular Age-Related Macular Degeneration: Early Detection of Neovascular Age-Related Macular Degeneration Study.

PURPOSE: To evaluate the diagnostic accuracy of routinely used tests of visual function and retinal morphology compared with fundus fluorescein angiography (FFA) to detect onset of active macular neovascularization in unaffected fellow eyes of patients with unilateral neovascular age-related macular degeneration (nAMD). DESIGN: Prospective diagnostic accuracy cohort study conducted in 24 eye clinics in the United Kingdom over 3 years. PARTICIPANTS: Older adults (>50 years) with recently diagnosed unilateral nAMD with a fellow (study) eye free of nAMD. METHODS: Self-reported vision, Amsler, clinic-measured visual acuity (VA), fundus assessment, and spectral domain OCT. The reference standard is FFA. MAIN OUTCOME MEASURES: Sensitivity and specificity of the 5 index tests. RESULTS: Of 552 participants monitored for up to 3 years, 145 (26.3%) developed active nAMD in the study eye, of whom 120 had an FFA at detection and constituted the primary analysis cohort. Index test positives at nAMD detection in those confirmed by FFA were self-reported vision much worse (5), distortion on Amsler (33), 10-letter decrease in acuity from baseline (36), fundus examination (64), and OCT (110). Percentage index test sensitivities were: self-reported vision 4.2 (95% confidence interval [CI], 1.6-9.8); Amsler 33.7 (95% CI, 25.1-43.5); VA 30.0 (95% CI, 22.5-38.7); fundus examination 53.8 (95% CI, 44.8-62.5); and OCT 91.7 (95% CI, 85.2-95.6). All 5 index test specificities were high at 97.0 (95% CI, 94.6-98.5), 81.4 (95% CI, 76.4-85.5), 66.3 (95% CI, 61.0-71.1), 97.6 (95% CI, 95.3-98.9), and 87.8 (95% CI, 83.8-90.9), respectively. The combination of OCT with one other index test that was a secondary outcome measure increased sensitivity marginally and decreased specificity for all combinations except fundus examination. CONCLUSIONS: Tests of self-reported change in vision, unmasking of new distortion, measurements of acuity, and fundus checks to diagnose active nAMD performed poorly in contrast to OCT. Our findings support a change to guidelines in clinical practice to monitor for onset of nAMD.

Corneal neovascularization.

The optical clarity of the cornea is essential for maintaining good visual acuity. Corneal neovascularization, which is a major cause of vision loss worldwide, leads to corneal opacification and often contributes to a cycle of chronic inflammation. While numerous factors prevent angiogenesis within the cornea, infection, inflammation, hypoxia, trauma, corneal degeneration, and corneal transplantation can all disrupt these homeostatic safeguards to promote neovascularization. Here, we summarize its etiopathogenesis and discuss the molecular biology of angiogenesis within the cornea. We then review the clinical assessment and diagnostic evaluation of corneal neovascularization. Finally, we describe current and emerging therapies.

Novel Parameters to Assess the Severity of Corneal Neovascularization Using Anterior Segment Optical Coherence Tomography Angiography.

PURPOSE: Assessment of anterior segment-optical coherence tomography angiography (AS-OCTA) to determine severity of corneal neovascularization (CoNV). DESIGN: Retrospective, cross-sectional, single-center study. METHODS: Patients of various CoNV etiologies were selected and classified into mild, moderate, and severe. Their AS-OCTA images were measured for CoNV anterior limit, CoNV posterior limit, CoNV thickness, CoNV depth%, CoNV vessel density, CoNV area, and CoNV volume. Further, AS-OCTA parameters were correlated to clinical parameters, such as classification, a numerical severity scale, vascular clock hours, and best-corrected visual acuity (BCVA). RESULTS: A total of 19 mild, 10 moderate, and 6 severe CoNV eyes were included with no significant age-gender differences. CoNV depth% and volume increased from mild to moderate (9.3 ± 1.1% to 17.7 ± 3.3%, P = .030, and 0.2 ± 0.1 mm3 to 1.0 ± 0.3 mm3, P = .025, respectively) and from moderate to severe CoNV (44.6 ± 5.3%, P < .001, and 2.0 ± 0.3 mm3, P = .014, respectively). CoNV area and posterior limit increased from mild to moderate (1.7 ± 0.3 mm2 to 4.6 ± 0.7 mm2, P = .001, and 217.7 ± 16.8 µm to 349.1 ± 54.9 µm, P = .048, respectively), not from moderate to severe (P = .999 and P = .403, respectively). CoNV thickness increased from moderate to severe (218.2 ± 46.6 µm to 340.2 ± 8.7 µm, P = .020), but not from mild to moderate. CoNV area and volume showed good correlations to CoNV staging (r = 0.703 and r = 0.771, respectively; P < .001) and severity scale (r = 0.794 and r = 0.712, respectively; P < .001). CoNV area showed good correlation to clock hours (r = 0.749, P < .001). CoNV depth and volume showed good correlation to BCVA (r = 0.744 and r = 0.722, respectively; P < .001). CoNV anterior limit and vessel density showed no significant correlations (P ≥ .05). CONCLUSIONS: Severe CoNV shows greater CoNV posterior limit, thickness, depth%, area, and volume on AS-OCTA compared to mild. CoNV volume and depth strongly correlate to BCVA. AS-OCTA provides novel, quantitative, and noninvasive parameters for assessing CoNV severity.

Treating corneal neovascularization using a combination of anti-VEGF injection and argon laser photocoagulation application - case report.

The present study aimed to demonstrate the possibility to treat corneal neovascularization using the combination of anti-VEGF injection and argon laser photocoagulation.

Comparison of Topical Pigment Epithelium-Derived Factor (PEDF) with Topical Bevacizumab for Accelerating the Regression of Corneal Neovascularization in an Experimental Model of Rabbit Corneal Angiogenesis.

Purpose: To evaluate the anti-angiogenic effect of topical administration of Pigment epithelium-derived factor (PEDF) on the reduction of corneal neovascularization (NV) in comparison to topical Bevacizumab.Methods: 18 eyes of 18 New Zealand rabbits were enrolled. Corneal NV was induced by a 7-0 silk suture. After suture removal, rabbits were randomly divided into three groups. In every group, one eye randomly treated with topical bevacizumab or topical PEDF or saline for 14 days. The area and length of neovascularization were measured by Image J. Histological studies were done in three groups.Results: After 14 days, the mean decrease of corneal NV length was 1.84 ± 0.17 mm (P < .001) in PEDF group and 1.6 ± 0.07 mm (P < .001) in bevacizumab group which was significantly more than the saline group (P = .001 and P < .001, respectively). There was no significant difference between PEDF and bevacizumab group in the reduction of corneal NV length (P = .85). The mean decrease of corneal NV area was 4.94 ± 0.55 mm2 (P < .001) in PEDF group and 4.23 ± 0.29 mm2 in the bevacizumab group (P < .001). PEDF and bevacizumab significantly decreased corneal NV area in comparison to the saline group (p = .017, p = .001, respectively). The mean decrease of corneal NV area did not show a significant difference between PEDF and bevacizumab groups (P = .72).Conclusion: Topical PEDF might be an effective and safe treatment option as bevacizumab in a short-term use, indicating that it is as good as the standard. However, long-term effect is required to be investigated.

Early Application of Bevacizumab After Sclerocorneal Grafting for Patients With Severe Late-Stage Ocular Chemical Burns.

PURPOSE: To investigate whether subconjunctival bevacizumab help prevent corneal graft neovascularization and prolong the graft survival of patients with chemical burns. METHODS: We performed a prospective nonrandomized comparative case series study. Twenty-six eyes received subconjunctival bevacizumab (10 mg/0.4 mL) once and topical immunosuppressive agents after sclerocorneal lamellar keratoplasty as the treatment, and 13 eyes received a topical immunosuppressant alone and served as the control group. The main outcomes were a cumulative probability of graft survival, development of corneal neovascularization, and complications. RESULTS: The postoperative follow-up time was 14.3 months (range, 2-62 mo). The cumulative graft survival time was significantly longer in the treatment group than that in the control group (42.9 ± 5.9 vs. 4.8 ± 0.7 mo; log rank < 0.001). In the treatment group, 19 of the 26 grafts (73.1%) survived as transparent with a mean follow-up of 18.7 ± 3.0 months. At the end of the follow-up, 4 grafts remained free of neovascularization, 2 developed edema without neovascularization, and 15 remained transparent with a stable ocular surface and some neovascular vessels in the peripheral transplant interface. The other 5 grafts became opaque and neovascularized. In the control group, all grafts became opaque and neovascularized within the follow-up period (5.5 ± 0.7 mo). During the follow-up, a corneal epithelial defect developed in 9 eyes in the treatment group and 7 in the control group. CONCLUSIONS: Early application of subconjunctival bevacizumab after sclerocorneal lamellar keratoplasty can significantly prevent corneal neovascularization and promote graft survival for severe late-stage ocular chemical burns.

Intrastromal bevacizumab in the management of corneal neovascularization: a retrospective review.

PURPOSE: To evaluate the long-term safety and efficacy of intrastromal bevacizumab for treatment of deep corneal neovascularization in candidates for high-risk cornea grafting. METHODS: A single-center retrospective study involving 14 eyes of 14 patients with chronic deep corneal neovascularization, treated with intrastromal bevacizumab by a single provider from 2011 to present. Intrastromal bevacizumab (0.05-0.1 mL of 2.5 mg/0.1 mL) was administered every 4-8 weeks. On average 1-3 intrastromal injections were performed prior to corneal grafting (penetrating keratoplasty or deep anterior lamellar keratoplasty). RESULTS: 64.2% patients had neurotrophic keratitis secondary to herpes zoster or simplex. Neovascularization was encroaching the visual axis in 50% and was paracentral in 42.8%. After intrastromal bevacizumab injection, 14.2% had complete regression of neovascularization, avoiding the need of future corneal transplant. Persistent neovascularization was noticed in 21.4%. Successful penetrating keratoplasty was performed in 57% of patients. Minimal adverse effects were noted; temporary epithelial defect was seen in two eyes and self-limited intrastromal hemorrhage in one. There was no evidence of recurrence of neovascularization or graft rejection in the transplant group (mean follow-up 3 years). CONCLUSION: Intrastromal bevacizumab appears to be a safe and effective modality in the treatment of chronic corneal neovascularization, producing durable regression of corneal neovascularization and allowing for durable success of subsequent corneal transplants in high-risk patients.

Corneal sterile infiltration induced by topical use of ocular hypotensive agent.

PURPOSE: To report two cases with corneal sterile infiltration presumably due to topical ocular hypotensive agent. METHOD: Case report. RESULTS: Case 1: A 65-year-old man presented with corneal opacity and neovascularization in his left eye. A diagnosis of glaucoma was made 2 years previously, and anti-glaucoma agents were prescribed (brimonidine tartrate, ripasudil hydrochloride hydrate, and brinzolamide) for both eyes. Case 2: A 75-year-old woman noticed corneal opacity in the left eye. A diagnosis of glaucoma was made 35 years previously, and anti-glaucoma agents were prescribed (brimonidine tartrate, 1% dorzolamide, and bimatoprost) for both eyes. In both cases, ocular examination revealed follicular conjunctivitis and blepharitis in both eyes, and corneal sterile infiltration with neovascularization in the left eyes. The three topical drugs were discontinued and replaced with 0.1% fluorometholone. Both the blepharitis and corneal sterile infiltration improved thereafter, although corneal opacity remained across the stromal layer. CONCLUSION: We encountered two cases of corneal and conjunctival complications that were suspected as side effects after brimonidine eye drop use. Special care should be taken to observe the condition of ocular surface when topical brimonidine is administered.

Application of Corneal Optical Coherence Tomography Angiography for Assessment of Vessel Depth in Corneal Neovascularization.

PURPOSE: To map and measure the depths of corneal neovascularization (NV) using 3-dimensional optical coherence tomography angiography (OCTA) at 2 different wavelengths. METHODS: Corneal NV of varying severity, distribution, and underlying etiology was examined. Average NV depth and vessel density were measured using 840-nm spectral-domain OCTA and 1050-nm swept-source OCTA. The OCTA results were compared with clinical slit-lamp estimation of NV depth. RESULTS: Twelve eyes with corneal NV from 12 patients were imaged with OCTA. Clinically "superficial," "midstromal," and "deep" cases had an average vessel depth of 23%, 39%, and 66% on 1050-nm OCTA, respectively. Average vessel depth on OCTA followed a statistically significant ordinal trend according to the clinical classification of vessel depth (Jonckheere-Terpstra test, P < 0.001). In 8 cases where both 840-nm OCTA and 1050-nm OCTA were acquired, there was excellent agreement in the mean vessel depth between the 2 systems (concordance correlation coefficient = 0.94, P < 0.001). The average vessel density measured by 840-nm OCTA was higher (average 1.6-fold) than that measured by 1050-nm OCTA. CONCLUSIONS: Corneal OCTA was able to map corneal NV in 3 dimensions and measure vessel depth and density. The depth of corneal NV varied between different pathologies in a manner consistent with previous pathologic studies. The measured vessel density appeared to be affected by the interscan time, which affects blood flow velocity sensitivity, and the wavelength, which affects the ability to penetrate through opacity. These findings suggest possible clinical applications of OCTA for the diagnosis of corneal pathology and quantitative monitoring of therapeutic response in patients with corneal NV.