Artificial intelligence for diagnosing exudative age-related macular degeneration.

BACKGROUND: Age-related macular degeneration (AMD) is a retinal disorder characterized by central retinal (macular) damage. Approximately 10% to 20% of non-exudative AMD cases progress to the exudative form, which may result in rapid deterioration of central vision. Individuals with exudative AMD (eAMD) need prompt consultation with retinal specialists to minimize the risk and extent of vision loss. Traditional methods of diagnosing ophthalmic disease rely on clinical evaluation and multiple imaging techniques, which can be resource-consuming. Tests leveraging artificial intelligence (AI) hold the promise of automatically identifying and categorizing pathological features, enabling the timely diagnosis and treatment of eAMD. OBJECTIVES: To determine the diagnostic accuracy of artificial intelligence (AI) as a triaging tool for exudative age-related macular degeneration (eAMD). SEARCH METHODS: We searched CENTRAL, MEDLINE, Embase, three clinical trials registries, and Data Archiving and Networked Services (DANS) for gray literature. We did not restrict searches by language or publication date. The date of the last search was April 2024. SELECTION CRITERIA: Included studies compared the test performance of algorithms with that of human readers to detect eAMD on retinal images collected from people with AMD who were evaluated at eye clinics in community or academic medical centers, and who were not receiving treatment for eAMD when the images were taken. We included algorithms that were either internally or externally validated or both. DATA COLLECTION AND ANALYSIS: Pairs of review authors independently extracted data and assessed study quality using the Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) tool with revised signaling questions. For studies that reported more than one set of performance results, we extracted only one set of diagnostic accuracy data per study based on the last development stage or the optimal algorithm as indicated by the study authors. For two-class algorithms, we collected data from the 2x2 table whenever feasible. For multi-class algorithms, we first consolidated data from all classes other than eAMD before constructing the corresponding 2x2 tables. Assuming a common positivity threshold applied by the included studies, we chose random-effects, bivariate logistic models to estimate summary sensitivity and specificity as the primary performance metrics. MAIN RESULTS: We identified 36 eligible studies that reported 40 sets of algorithm performance data, encompassing over 16,000 participants and 62,000 images. We included 28 studies (78%) that reported 31 algorithms with performance data in the meta-analysis. The remaining nine studies (25%) reported eight algorithms that lacked usable performance data; we reported them in the qualitative synthesis. Study characteristics and risk of bias Most studies were conducted in Asia, followed by Europe, the USA, and collaborative efforts spanning multiple countries. Most studies identified study participants from the hospital setting, while others used retinal images from public repositories; a few studies did not specify image sources. Based on four of the 36 studies reporting demographic information, the age of the study participants ranged from 62 to 82 years. The included algorithms used various retinal image types as model input, such as optical coherence tomography (OCT) images (N = 15), fundus images (N = 6), and multi-modal imaging (N = 7). The predominant core method used was deep neural networks. All studies that reported externally validated algorithms were at high risk of bias mainly due to potential selection bias from either a two-gate design or the inappropriate exclusion of potentially eligible retinal images (or participants). Findings Only three of the 40 included algorithms were externally validated (7.5%, 3/40). The summary sensitivity and specificity were 0.94 (95% confidence interval (CI) 0.90 to 0.97) and 0.99 (95% CI 0.76 to 1.00), respectively, when compared to human graders (3 studies; 27,872 images; low-certainty evidence). The prevalence of images with eAMD ranged from 0.3% to 49%. Twenty-eight algorithms were reportedly either internally validated (20%, 8/40) or tested on a development set (50%, 20/40); the pooled sensitivity and specificity were 0.93 (95% CI 0.89 to 0.96) and 0.96 (95% CI 0.94 to 0.98), respectively, when compared to human graders (28 studies; 33,409 images; low-certainty evidence). We did not identify significant sources of heterogeneity among these 28 algorithms. Although algorithms using OCT images appeared more homogeneous and had the highest summary specificity (0.97, 95% CI 0.93 to 0.98), they were not superior to algorithms using fundus images alone (0.94, 95% CI 0.89 to 0.97) or multimodal imaging (0.96, 95% CI 0.88 to 0.99; P for meta-regression = 0.239). The median prevalence of images with eAMD was 30% (interquartile range [IQR] 22% to 39%). We did not include eight studies that described nine algorithms (one study reported two sets of algorithm results) to distinguish eAMD from normal images, images of other AMD, or other non-AMD retinal lesions in the meta-analysis. Five of these algorithms were generally based on smaller datasets (range 21 to 218 participants per study) yet with a higher prevalence of eAMD images (range 33% to 66%). Relative to human graders, the reported sensitivity in these studies ranged from 0.95 and 0.97, while the specificity ranged from 0.94 to 0.99. Similarly, using small datasets (range 46 to 106), an additional four algorithms for detecting eAMD from other retinal lesions showed high sensitivity (range 0.96 to 1.00) and specificity (range 0.77 to 1.00). AUTHORS' CONCLUSIONS: Low- to very low-certainty evidence suggests that an algorithm-based test may correctly identify most individuals with eAMD without increasing unnecessary referrals (false positives) in either the primary or the specialty care settings. There were significant concerns for applying the review findings due to variations in the eAMD prevalence in the included studies. In addition, among the included algorithm-based tests, diagnostic accuracy estimates were at risk of bias due to study participants not reflecting real-world characteristics, inadequate model validation, and the likelihood of selective results reporting. Limited quality and quantity of externally validated algorithms highlighted the need for high-certainty evidence. This evidence will require a standardized definition for eAMD on different imaging modalities and external validation of the algorithm to assess generalizability.

SCORE2 Report 24: Nonlinear Relationship of Retinal Thickness and Visual Acuity in Central Retinal and Hemiretinal Vein Occlusion.

PURPOSE: To investigate whether a nonlinear association between central subfield thickness (CST) on spectral-domain OCT and concurrent visual acuity letter score (VALS) exists in eyes treated initially with aflibercept or bevacizumab for macular edema associated with central retinal vein occlusion (CRVO) or hemiretinal vein occlusion (HRVO) in the Study of Comparative Treatments for Retinal Vein Occlusion 2 (SCORE2). DESIGN: Long-term follow-up after a randomized clinical trial from 64 centers in the United States. PARTICIPANTS: Participants were followed up to 60 months and treated at investigator discretion after completing the 12-month treatment protocol. METHODS: Two-segment linear regression models were compared with simple linear regression models of VALS on CST. Pearson correlation coefficients were calculated to assess strength of CST and VALS associations. MAIN OUTCOME MEASURES: Central subfield thickness was measured by OCT and VALS by the electronic Early Treatment Diabetic Retinopathy Study methodology. RESULTS: Estimated inflection points, reflecting turning points at which the CST and VALS association changes from positive to negative, calculated at 7 postbaseline visits, range from 217 to 256 µm. A strongly positive correlation exists to the left of each estimated inflection point, ranging from 0.29 (P < 0.01 at month 60) to 0.50 (P < 0.01 at month 12), and a strongly negative correlation exists to the right of each estimated inflection point, ranging from -0.43 (P < 0.01 at month 1) to -0.74 (P < 0.01 at month 24). Randomization statistical tests showed that 2-segment models are favored over 1-segment models for all postbaseline months (P < 0.001 for all tests performed). CONCLUSIONS: The relationship between CST and VALS in eyes with CRVO or HRVO after treatment with anti-vascular endothelial growth factor (VEGF) therapy is not simply linear. The usually modest correlations between OCT-measured CST and visual acuity belie strong left and right correlations present in 2-segment models. Post-treatment CST close to the estimated inflection points showed the best expected VALS. The SCORE2 participants with a post-treatment CST after treatment close to the estimated inflection points of 217 to 256 µm showed the best VALS. In patients treated with anti-VEGF for macular edema associated with CRVO or HRVO, a thinner retina is not always associated with better VALS. FINANCIAL DISCLOSURE(S): Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Intravitreal acetazolamide implant for pseudophakic cystoid macular edema.

BACKGROUND: Pseudophakic cystoid macular edema (PCME) is the most common cause of visual acuity deterioration after uncomplicated cataract surgery. There is no consensus regarding how to manage recurrent or refractory cases. REPORT: A 54-year-old woman complained of decreased vision and central metamorphopsia in the right eye (OD) 3 months after uneventful cataract surgery. Visual acuity was 0.3 logMAR (20/40) OD and 0.1 logMAR (20/25) OS. Reduced macular brightness was seen OD on funduscopy associated with increased macular thickness on optical coherence tomography (OCT). Pseudophakic cystoid macular edema (PCME) was diagnosed, and treatment with oral acetazolamide was tried without success. The patient underwent a single intravitreal injection of an acetazolamide implant (260 µg) OD as off-label treatment. Four weeks following the injection, she reported complete resolution of her metamorphopsia and visual loss OD. Four months later, her visual acuity was 0.0 logMAR (20/20) in OD and 0.1 logMAR (20/25) in OS. The patient reported no discomfort after the injection procedure. Laboratory and ophthalmologic tests did not identify any adverse effects of the medication. CONCLUSION: We show that PCME refractory to conventional treatment improved after intravitreal acetazolamide implant injection. Further investigation is warranted to confirm these preliminary findings.

Interventions for acute non-arteritic central retinal artery occlusion.

BACKGROUND: Acute non-arteritic central retinal artery occlusion (CRAO) occurs as a sudden interruption of the blood supply to the retina and typically results in severe loss of vision in the affected eye. Although many therapeutic interventions have been proposed, there is no generally agreed upon treatment regimen. OBJECTIVES: To assess the effects of treatments for acute non-arteritic CRAO. SEARCH METHODS: We searched the Cochrane Central Register of Controlled Trials (CENTRAL) (which contains the Cochrane Eyes and Vision Trials Register) (2022, Issue 2); Ovid MEDLINE; Embase.com; PubMed; Latin American and Caribbean Health Sciences Literature Database (LILACS); ClinicalTrials.gov; and the World Health Organization (WHO) International Clinical Trials Registry Platform (ICTRP). We did not use any date or language restrictions in the electronic search for trials. We last searched the electronic databases on 15 February 2022. SELECTION CRITERIA: We included randomized controlled trials (RCTs) comparing any interventions with another treatment in participants with acute non-arteritic CRAO in one or both eyes.  DATA COLLECTION AND ANALYSIS: We used standard Cochrane methodology and graded the certainty of the body of evidence for primary (mean change in best-corrected visual acuity [BCVA]) and secondary (quality of life and adverse events) outcomes using the GRADE classification. MAIN RESULTS: We included six RCTs with 223 total participants with acute non-arteritic CRAO; the studies ranged in size from 10 to 84 participants. The included studies varied geographically: one in Australia, one in Austria and Germany, two in China, one in Germany, and one in Italy. We were unable to conduct any meta-analyses due to study heterogeneity. None of the included studies compared the same pair of interventions: 1) tissue plasminogen activator (t-PA) versus intravenous saline; 2) t-PA versus isovolemic hemodilution, eyeball massage, intraocular pressure reduction, and anticoagulation; 3) nitroglycerin, methazolamide, mecobalamin tablets, vitamin B1 and B12 injections, puerarin and compound anisodine (also known as 654-2) along with oxygen inhalation, eyeball massage, tube expansion, and anticoagulation compared with and without intravenous recombinant tissue plasminogen activator (rt-PA); 4) transcorneal electrical stimulation (TES) with 0 mA versus with 66% of the participant's individual electrical phosphene threshold (EPT) at 20 Hz (66%) versus with 150% of the participant's individual EPT (150%) at 20 Hz; 5) ophthalmic artery branch retrograde thrombolysis versus superselective ophthalmic artery thrombolysis; and 6) pentoxifylline versus placebo. There was no evidence of an important difference in visual acuity between participants treated with t-PA versus intravenous saline (mean difference [MD] at 1 month -0.15 logMAR, 95% confidence interval [CI] -0.48 to 0.18; 1 study, 16 participants; low certainty evidence); t-PA versus isovolemic hemodilution, eyeball massage, intraocular pressure reduction, and anticoagulation (MD at 1 month -0.00 logMAR, 95% CI -0.24 to 0.23; 1 study, 82 participants; low certainty evidence); and TES with 0 mA versus TES with 66% of EPT at 20 Hz versus TES with 150% of EPT at 20 Hz. Participants treated with t-PA experienced higher rates of serious adverse effects. The other three comparisons did not report statistically significant differences. Other studies reported no data on secondary outcomes (quality of life or adverse events).  AUTHORS' CONCLUSIONS: The current research suggests that proposed interventions for acute non-arteritic CRAO may not be better than observation or treatments of any kind such as eyeball massage, oxygen inhalation, tube expansion, and anticoagulation, but the evidence is uncertain. Large, well-designed RCTs are necessary to determine the most effective treatment for acute non-arteritic CRAO.

Virtual supervision in ophthalmology: a scoping review.

PURPOSE: The published information on virtual supervision (VS) in ophthalmology is not well described. This scoping review describes the evidence and potential role for VS in ophthalmic practice and education. METHODS: A literature search strategy was developed in accordance with Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR). We included full-text articles published in an English-language peer-reviewed journal that involved physician-physician or physician-trainee VS in ophthalmology. We excluded studies with direct (in-person) supervision. Two investigators independently extracted from each article the year of publication and study location, design, participant characteristics, sample size, and outcomes. We appraised the methodological quality of the studies using the Mixed Methods Appraisal Tool (MMAT). RESULTS: Seven articles were included in our qualitative synthesis. Supervisees ranged from physicians such as an ophthalmic surgeon and a general practitioner to medical trainees such as ophthalmology residents, vitreoretinal fellows, and emergency medicine residents. Study settings included emergency departments, operating rooms, eye clinics, and a rural hospital. All studies reported successful transmission of real-time images or videos of clinical examinations and surgical or in-office procedures. Various methods were used to ensure high image and video quality during VS, although some technical challenges remained. MMAT ratings revealed limitations in outcome measurement, statistical analysis, sampling strategy, and inclusion of confounding factors. CONCLUSION: Virtual supervision in ophthalmology is technologically feasible and permits synchronous communication and transmission of clinical data, which can be used to formulate diagnostic and management plans and learn new surgical skills. Future studies with larger sample sizes and robust study designs should investigate factors that make VS effective in ophthalmic practice and education.

Association of staffing with Incidence of delayed retinal break or detachment after posterior vitreous detachment in a resident urgent care clinic.

PURPOSE: To compare the incidence rate of delayed retinal break or detachment after diagnosis of acute, symptomatic posterior vitreous detachment (PVD) in a resident-run urgent care clinic (UCC) when staffed by a retina attending, non-retina ophthalmology attending, optometrist, or ophthalmology resident only. METHODS: Retrospective consecutive case series. Of the 594 patients with acute, symptomatic PVD evaluated in the UCC at Penn State Eye Center between 1/1/2016 and 10/10/2019, 454 were included in the study; 140 were excluded because they were diagnosed with a retinal break or detachment on presentation to the UCC, had media opacity precluding examination, or had no follow-up within one year. Demographics, presenting examination findings, and type of staffing were recorded; subsequent visits up to 1 year were analyzed for presence of delayed retinal break or detachment. RESULTS: Among 491 eyes of 454 patients with a mean follow-up of 147 days, ten delayed breaks (10/491, 2.0%) and three delayed detachments (3/491, 0.6%) were discovered. Incidence rates of delayed breaks and detachments were 1.8% (5/282) and 0.7% (2/282), respectively, in the retina attending group, 1.0% (1/105) and 1.0% (1/105) in the non-retina ophthalmology attending group, 4.7% (3/64) and 0% (0/64) in the optometrist group, and 2.5% (1/40) and 0% (0/40) in the ophthalmology resident only group. There was no statistically significant difference in the incidence of delayed break or detachment among the staffing groups (P = 0.7312), but this study was underpowered to detect a statistically significant difference among staffing groups. Patients with a delayed break or detachment were more likely to have lattice degeneration (P = 0.0265) or a history of retinal break in the contralateral eye (P = 0.0014), and most eyes (10 [76.9%]) with a delayed break or detachment were left eyes (P = 0.0466). CONCLUSIONS: The overall rate of delayed retinal break or detachment in the current study is similar to previously published rates among retinal physician and retinal fellow examiners. Although no statistically significant difference among staffing groups in the incidence rates of delayed retinal tears or detachments was identified in the study, it is important to note that the optometry and ophthalmology resident only groups had higher incidence rates of delayed retinal breaks than did the retina and non-retina ophthalmology attending groups, and this may be clinically important. Larger cohort studies would be needed in order to have the power to detect statistically significant differences among staffing groups. Varied staffing for acute, symptomatic PVD may assist with resource allocation in similar settings.

SCORE2 Report 17: Macular thickness fluctuations in anti-VEGF-treated patients with central or hemiretinal vein occlusion.

PURPOSE: To evaluate macular thickness fluctuations and their association with visual acuity outcome in eyes with macular edema (ME) secondary to central (CRVO) or hemiretinal vein occlusion (HRVO) treated initially with intravitreal aflibercept or bevacizumab. METHODS: Post hoc analysis of 362 patients with ME secondary to CRVO or HRVO initially randomized to six monthly intravitreal injections of aflibercept or bevacizumab. Three spectral domain optical coherence tomography (SD-OCT) central subfield thickness (CST) fluctuation measures were investigated over Months 1-12: standard deviation (SD), number of turning points (T) for each participant, and a measure denoted as Zigzag reflecting the magnitude of alternating ups and downs in a participant's CST. Main outcome measure is Month 12 visual acuity letter score (VALS). RESULTS: More fluctuations occurred in eyes randomized to bevacizumab than aflibercept: SD (59.98 vs 32.12; p < 0.0001), T (4.03 vs 3.53; p = 0.02) and Zigzag (24.91 vs 11.60; p = 0.0003). Month 12 VALS is significantly lower for the 4th (highest) quartile of the CST fluctuation measure than for the 1st (lowest) quartile for both SD (mean difference in VALS of 7.87; 95% confidence interval: 3.03, 12.70) and Zigzag (mean difference in VALS of 5.11; 95% confidence interval: 0.29, 9.93). SD and Zigzag quartiles were no longer significantly different after Month 1 VALS was added to the regression analysis. CONCLUSIONS: Greater CST fluctuation as assessed by SD and Zigzag was negatively associated with Month 12 VALS. However, early post-treatment VALS is a stronger predictor of VALS outcomes than the CST fluctuation measures.

Retinal vascular occlusions.

Acute retinal vascular occlusions are common causes of visual impairment. Although both retinal artery occlusions and retinal vein occlusions are associated with increased age and cardiovascular risk factors, their pathophysiology, systemic implications, and management differ substantially. Acute management of retinal artery occlusions involves a multidisciplinary approach including neurologists with stroke expertise, whereas treatment of retinal vein occlusions is provided by ophthalmologists. Optimisation of systemic risk factors by patients' primary care providers is an important component of the management of these two disorders.

Proliferative diabetic retinopathy treated with intravitreal ranibizumab and photocoagulation directed at ischemic retinal areas-A randomized study.

PURPOSE: To compare ETDRS panretinal laser photocoagulation (PRP) combined with intravitreal injection of ranibizumab (IVR) and photocoagulation targeted to ischemic retina (PIR) combined with IVR in patients with proliferative diabetic retinopathy (PDR). METHODS: PDR patients were randomly assigned to treatment with either PRP + IVR or PIR + IVR. ETRDS Best-corrected visual acuity (BCVA) and central subfield thickness (CSFT) measured on optic-coherence tomography images (OCT-Heidelberg Spectralis) were recorded at baseline and every 4 weeks for one year. Fluorescein leakage area (FLA) from active new vessels was measured every 12 weeks. Full-field ERG was recorded by means of DTL electrodes, following ISCEV standard recommendations, at baseline and after 3 months. RESULTS: Twenty-eight eyes completed the study period. At baseline, mean ± SE BCVA (logMAR) was 0.44 ± 0.07 and 0.37 ± 0.08 (P = 0.5030); CSFT (µm) was 324.0 ± 20.4 and 330.1 ± 22.1 (P = 0.8417); and FLA (mm2) was 16.10 ± 4.42 and 9.97 ± 1.83 (P = 0.2114) for PRP + IVR and PIR + IVR groups, respectively. There were no relevant changes on BCVA or CSFT, but a significant reduction for FLA was observed at all visits compared to baseline for both groups, with no differences between groups. ERG showed at baseline reduced dark-adapted amplitudes, and these changes were also significantly amplified after laser treatment. ROD b-wave amplitude was further reduced in 62 ± 6% for PRP + IVR and 59 ± 4% for group PIR + IVR, but with no between-groups significant difference (P = 0.9082). CONCLUSIONS: PIR + IVR or PRP + IVR are comparable strategies regarding FLA control in PDR and led to similar retinal function impairment, based on ERG changes up to one-year follow-up. TRIAL REGISTRATION NUMBER: NCT03904056, date of registration 02/11/2019, retrospectively registered.

Virtual reality training for cataract surgery operating performance in ophthalmology trainees.

BACKGROUND: Cataract surgery is the most common incisional surgical procedure in ophthalmology and is important in ophthalmic graduate medical education. Although most ophthalmology training programs in the United States (US) include virtual reality (VR) training for cataract surgery, comprehensive reviews that detail the impact of VR training on ophthalmology trainee performance are lacking. OBJECTIVES: To assess the impact of VR training for cataract surgery on the operating performance of postgraduate ophthalmology trainees, measured by operating time, intraoperative complications, postoperative complications, supervising physician ratings, and VR simulator task ratings. SEARCH METHODS: We searched CENTRAL (which contains the Cochrane Eyes and Vision Trials Register), Ovid MEDLINE, Embase.com, PubMed, LILACS, ClinicalTrials.gov, and the World Health Organization (WHO) International Clinical Trials Registry Platform (ICTRP). We did not use any date or language restrictions in the electronic search for trials. We last searched the electronic databases on 14 June 2021. SELECTION CRITERIA: We included randomized controlled trials (RCTs) comparing VR training to any other method of training, including non-VR simulation training (e.g., wet laboratory training), didactics training, or no supplementary training in postgraduate ophthalmology trainees. DATA COLLECTION AND ANALYSIS: We used standard Cochrane methodology. Primary outcomes were operating times in the operating room and intraoperative complications. Secondary outcomes were operating times in simulated settings, simulator task ratings, and supervising physician ratings, either in the operating room or simulated settings. MAIN RESULTS: We included six RCTs with a total of 151 postgraduate ophthalmology trainees ranging from 12 to 60 participants in each study. The included studies varied widely in terms of geography: two in the US, and one study each in China, Germany, India, and Morocco. Three studies compared VR training for phacoemulsification cataract surgery on the Eyesi simulator (VRmagic, Mannheim, Germany) with wet laboratory training and two studies compared VR training with no supplementary training. One study compared trainees who received VR training with those who received conventional training for manual small incision cataract surgery on the HelpMeSee simulator (HelpMeSee, New York, NY). Industry financially supported two studies. All studies had at least three domains judged at high or unclear risks of bias. We did not conduct a meta-analysis due to insufficient data (i.e., lack of precision measurements, or studies reported only P values). All evidence was very low-certainty, meaning that any estimates were unreliable. The evidence for the benefits of VR training for trainees was very uncertain for primary outcomes. VR-trained trainees relative to those without supplementary training had shorter operating times (mean difference [MD] -17 minutes, 95% confidence interval [CI] -21.62 to -12.38; 1 study, n = 12; very low-certainty evidence). Results for operating time were inconsistent when comparing VR and wet laboratory training: one study found that VR relative to wet laboratory training was associated with longer operating times (P = 0.038); the other reported that two training groups had similar operating times (P = 0.14). One study reported that VR-trained trainees relative to those without supplementary training had fewer intraoperative complications (P < 0.001); in another study, VR and conventionally trained trainees had similar intraoperative complication rates (MD -8.31, 95% CI -22.78 to 6.16; 1 study, n = 19; very low-certainty evidence). For secondary outcomes, VR training may have similar impact on trainee performance compared to wet laboratory and greater impact compared to no supplementary training, but the evidence was very uncertain. One study reported VR-trained trainees relative to those without supplementary training had significantly reduced operating time in simulated settings (P = 0.0013). Another study reported that VR-trained relative to wet laboratory-trained trainees had shorter operating times in VR settings (MD -1.40 minutes, 95% CI -1.96 to -0.84; 1 study, n = 60) and similar times in wet laboratory settings (MD 0.16 minutes, 95% CI -0.50 to 0.82; 1 study, n = 60). This study also found the VR-trained trainees had higher VR simulator ratings (MD 5.17, 95% CI 0.61 to 9.73; 1 study, n = 60). Results for supervising physician ratings in the operating room were inconsistent: one study reported that VR- and wet laboratory-trained trainees received similar supervising physician ratings for cataract surgery (P = 0.608); another study reported that VR-trained trainees relative to those without supplementary training were less likely to receive poor ratings by supervising physicians for capsulorhexis construction (RR 0.29, 95% CI 0.15 to 0.57). In wet laboratory settings, VR-trained trainees received similar supervising physician ratings compared with wet laboratory-trained trainees (MD -1.50, 95% CI -6.77 to 3.77; n = 60) and higher supervising physician ratings compared with trainees without supplementary training (P < 0.0001). However, the results for all secondary outcomes should be interpreted with caution because of very low-certainty evidence.  AUTHORS' CONCLUSIONS: Current research suggests that VR training may be more effective than no supplementary training in improving trainee performance in the operating room and simulated settings for postgraduate ophthalmology trainees, but the evidence is uncertain. The evidence comparing VR with conventional or wet laboratory training was less consistent.

Association between early and late response in eyes with central or hemiretinal vein occlusion treated with anti-VEGF agents : SCORE2 report 12: secondary analysis of the SCORE2 clinical trial.

PURPOSE: To assess whether early visual acuity letter score change from baseline (ΔVALS) and early spectral domain optical coherence tomography (SD-OCT) measures of center point thickness (CPT) are associated with later ΔVALS in eyes with macular edema due to central or hemiretinal vein occlusion treated with intravitreal aflibercept or bevacizumab. METHODS: Secondary analysis of a randomized clinical trial of 362 participants. RESULTS: Considered separately at month 3, CPT (categorized as ≤ 300 µm, > 300 µm) and ΔVALS (categorized as < 5, 5-9, ≥ 10) are predictive of ΔVALS at month 6 (aflibercept: P = 0.02 for CPT and P < 0.0001 for ΔVALS; bevacizumab: P = 0.007 for CPT and P < 0.0001 for ΔVALS) and, except for CPT in the bevacizumab arm, also predictive of ΔVALS at month 12 (aflibercept: P = 0.03 for CPT and P < 0.0001 for ΔVALS; bevacizumab: P = 0.18 for CPT and P < 0.0001 for ΔVALS). Month 3 predictors are also associated with average ΔVALS from months 4 to 12 (CPT P = 0.01 in the aflibercept arm, P = 0.02 in the bevacizumab arm; ΔVALS > 10 versus < 5; P < 0.001 for both aflibercept and bevacizumab). When month 3 measures are considered jointly, ΔVALS effect remains significant for average ΔVALS from months 4 to 12 (aflibercept: P = 0.002; bevacizumab: P < 0.0001) but not CPT (aflibercept: P = 0.18; bevacizumab: P = 0.22). CONCLUSION: While both month 3 ΔVALS and CPT are predictive of ΔVALS after month 3 through month 12, early ΔVALS has a stronger relationship than CPT with later ΔVALS. SCORE2 registration number is NCT01969708.

Combined pars plana vitrectomy (PPV) and phacoemulsification (phaco) versus PPV and deferred phaco for phakic patients with full-thickness macular hole (FTMH) and no significant cataract at baseline: 1-year outcomes of a randomized trial combined PPV/phaco vs PPV/deferred phaco for MH.

PURPOSE: To compare functional and anatomic outcomes of combined pars plana vitrectomy (PPV) and phacoemulsification (phaco) versus PPV and deferred phaco in patients with full-thickness macular hole (FTMH) and no significant cataract. METHODS: Thirty-four patients were randomized to group 1 (combined PPV/phaco) and 34 to group 2 (PPV/deferred phaco). Group 2 patients could undergo phaco any time after FTMH surgery if significant cataract developed. RESULTS: Sixty-five patients (33 group 1 and 32 group 2) completed the 12-month visit. Mean ± SEM logMAR best-corrected visual acuity (BCVA) was 0.92 ± 0.04 and 0.90 ± 0.04 at baseline and improved significantly to 0.60 ± 0.05 and 0.58 ± 0.05 at month 12 (p < 0.0001) in groups 1 and 2, respectively. There was no significant difference between the groups in mean BCVA at baseline or at month 12. Mean macular sensitivity (dB) was 18.22 ± 0.93 and 16.72 ± 0.93 at baseline and increased to 21.13 ± 0.86 and 21.07 ± 0.85 in groups 1 and 2, respectively (p < 0.05) with no significant difference between the groups (p = 0.449) at month 12. FTMH closure rate was 73% and 75% in groups 1 and 2, respectively (p = 0.834). CONCLUSION: Among patients with FTMH and no significant cataract at baseline, combined PPV/phaco was associated with similar BCVA, microperimetry, and FTMH closure outcomes at 1-year compared with PPV/deferred phaco. TRIAL REGISTRATION: ( clinicaltrials.gov.br ): Ensaios clínicos brasileiros: RBR-3wmd9s; UTN number: U1111-1190-5013; Plataforma Brasil CAAE number: 50455415.3.0000.5440; IRB number: 1.433.000.

Bevacizumab versus triamcinolone for persistent diabetic macular edema: a randomized clinical trial.

PURPOSE: To evaluate 24-week visual acuity and anatomic outcomes of two "pro re nata" (prn) treatment strategies (intravitreal bevacizumab [IVB] prn versus intravitreal triamcinolone acetonide [IVT] prn) in patients with persistent diabetic macular edema (pDME) after 24 weeks of prn-IVB. METHODS: One hundred eyes with center-involving DME were enrolled and treated with prn-IVB for 24 weeks; at week 24, eyes with pDME (central subfield thickness [CST] on spectral domain optical coherence tomography > 300 µm) were randomized to IVB monthly prn (group I; prn-IVB) or IVT every 3 months prn (group II; prn-IVT) and eyes in which the CST was ≤ 300 µm were assigned to continue prn-IVB (group III). RESULTS: Seventy-four eyes completed a 48-week study period. At week 24, 65 (79.3%) eyes still had DME with CST > 300 µm and, therefore, were randomized to prn-IVB (group I, n = 33) or prn-IVT (group II, n = 32); the remaining 17 (20.7%) eyes had CST ≤ 300 µm and were assigned to continued treatment with prn-IVB (group III). At baseline, mean CST (µm) ± standard error of the mean (SEM) was 447.2 ± 24.4, 478.0 ± 19.7, and 386.0 ± 21.0 in groups I, II, and III, respectively (p > 0.05). At week 48, there was no significant difference in mean CST between groups I and II (369.9 ± 23.3 and 426.0 ± 26.1, respectively; p = 0.9995). A significant reduction in mean CST, compared with baseline, was noted at weeks 28 (p = 0.0002) and 44 (p = 0.0002) in group II. Group I did not show a significant reduction in mean CST compared with baseline at any study visit. There were no significant differences in mean CST between groups I and II at any study visit. At baseline, mean ± SEM best-corrected visual acuity (BCVA) (logMAR) was 0.50 ± 0.00, 0.60 ± 0.10, and 0.50 ± 0.10 in groups I, II, and III, respectively (p > 0.05). At week 48, there was no statistically significant difference in mean BCVA between groups I and II (0.50 ± 0.10 and 0.80 ± 0.10, respectively; p = 0.4473). There was no significant improvement in mean BCVA, as compared with baseline, at any study follow-up visit in any of the groups. Group II demonstrated significantly lower BCVA after 24 weeks of IVT (at week 48) compared with baseline (p = 0.0435). There was no significant difference in mean BCVA between groups I and II at any time-point. CONCLUSION: In eyes with pDME after 24 weeks of treatment with prn-IVB, there was no difference between continued treatment with prn-IVB versus a treatment switch to prn-IVT with respect to mean BCVA or mean CST at week 48. However, BCVA was stable in the prn-IVB group, while prn-IVT was associated with BCVA reduction from baseline and a higher risk of IOP elevation.

LONG-TERM PHYSICAL STABILITY, STERILITY, AND ANTI-VEGF BIOACTIVITY OF REPACKAGED BEVACIZUMAB IN 2-ML GLASS VIALS.

PURPOSE: Repackaged bevacizumab in single-dose, prefilled syringes for intravitreal injection is available, but with shelf life limited from 60 days to 90 days. For the Study of COmparative Treatments for REtinal Vein Occlusion 2 (SCORE2), 2-mL sterile glass vials were used rather than prefilled syringes to provide a longer shelf life for study supplies. METHODS: Repackaged bevacizumab in glass vials was tested at release and, for 1 lot, after 1, 3, 6, and 12 months for physical stability, including concentration, purity and appearance, and for sterility and endotoxins. Vials from 2 lots were tested at release and after 20 months and 21 months, respectively. One lot was tested at 21 months for anti-VEGF bioactivity compared with a fresh supply of commercial bevacizumab. RESULTS: Repackaged bevacizumab in 2-mL glass vials continued to meet all quality release specifications and remain sterile for up to 21 months. In addition, no degradation in anti-VEGF bioactivity was observed at 21 months compared with a fresh bevacizumab control. CONCLUSION: Bevacizumab can be repackaged into small, single-dose glass vials for intravitreal injection and the qualities of the commercial product maintained, including anti-VEGF bioactivity, for up to 21 months in refrigerated storage. Consideration should be given to repackaging bevacizumab for ophthalmic use in small glass vials as opposed to plastic syringes.

SCORE2 Report 2: Study Design and Baseline Characteristics.

PURPOSE: To describe the design and baseline characteristics of participants in the Study of COmparative Treatments for REtinal Vein Occlusion 2 (SCORE2) and to compare with cohorts from other retinal vein occlusion trials. DESIGN: Phase III prospective, multicenter, randomized clinical trial designed to assess whether intravitreal bevacizumab is noninferior to intravitreal aflibercept for treatment of decreased vision attributable to macular edema associated with central retinal vein occlusion (CRVO) or hemiretinal vein occlusion (HRVO). PARTICIPANTS: Total of 362 participants: 307 with CRVO and 55 with HRVO. METHODS: Demographic and study eye characteristics are summarized and compared between CRVO and HRVO study participants. MAIN OUTCOME MEASURES: Baseline ophthalmic characteristics, including visual acuity and retinal thickness, and medical history characteristics, including hypertension, diabetes mellitus, and coronary artery disease. RESULTS: The mean age of participants was 69 years, 76% of participants were white, and 90% were non-Hispanic. There was a racial disparity with respect to disease type, with 38% of HRVO patients being black compared with 11% of CRVO patients (P value adjusted for multiple testing = 0.0001). This is similar to findings from the previous SCORE Study. Comorbidities included hypertension (77%), diabetes mellitus (31%), and coronary artery disease (15%). At baseline, mean visual acuity letter score was 50 (20/100) (range, 19-73 [20/400 to 20/40]), mean optical coherence tomography (OCT)-measured central subfield thickness was 678 µm (range, 300-1203 µm), and mean number of months from diagnosis of macular edema to randomization was 6 (range, 0-104 months). One hundred twenty (33%) SCORE2 participants had been treated previously with anti-vascular endothelial growth factor (anti-VEGF) therapy, with these participants having baseline visual acuity letter score and OCT-measured central subfield thickness similar to those without prior anti-VEGF treatment, but longer mean duration of macular edema before randomization (18 months vs. 1 month for those without prior anti-VEGF treatment; P < 0.0001). CONCLUSIONS: The SCORE2 cohort is a heterogeneous population, including both CRVO and HRVO eyes and both treatment-naïve eyes and eyes treated previously with anti-VEGF, which will allow study results to have broad applicability to CRVO and HRVO patients receiving treatment for macular edema. Similarities of the baseline characteristics of the SCORE2 population to other CRVO trial cohorts will allow meaningful comparisons of outcome results across trials.

Effect of Bevacizumab vs Aflibercept on Visual Acuity Among Patients With Macular Edema Due to Central Retinal Vein Occlusion: The SCORE2 Randomized Clinical Trial.

IMPORTANCE: Studies have established the efficacy and safety of aflibercept for the treatment of macular edema due to central retinal vein occlusion. Bevacizumab is used off-label to treat this condition despite the absence of supporting data. OBJECTIVE: To investigate whether bevacizumab is noninferior to aflibercept for the treatment of macular edema secondary to central retinal or hemiretinal vein occlusion. DESIGN, SETTING, AND PARTICIPANTS: The SCORE2 randomized noninferiority clinical trial was conducted at 66 private practice or academic centers in the United States, and included 362 patients with macular edema due to central retinal or hemiretinal vein occlusion who were randomized 1:1 to receive aflibercept or bevacizumab. The first participant was randomized on September 17, 2014, and the last month 6 visit occurred on May 6, 2016. Analyses included data available as of December 30, 2016. INTERVENTIONS: Eyes were randomized to receive intravitreal injection of bevacizumab (1.25 mg; n = 182) or aflibercept (2.0 mg; n = 180) every 4 weeks through month 6. MAIN OUTCOMES AND MEASURES: The primary outcome was mean change in visual acuity (VA) letter score (VALS) from the randomization visit to the 6-month follow-up visit, based on the best-corrected electronic Early Treatment Diabetic Retinopathy Study VALS (scores range from 0-100; higher scores indicate better VA). The noninferiority margin was 5 letters, and statistical testing for noninferiority was based on a 1-sided 97.5% confidence interval. RESULTS: Among 362 randomized participants (mean [SD] age, 69 [12] years; 157 [43.4%] women; mean [SD] VALS at baseline, 50.3 [15.2] [approximate Snellen VA 20/100]), 348 (96.1%) completed the month 6 follow-up visit. At month 6, the mean VALS was 69.3 (a mean increase from baseline of 18.6) in the bevacizumab group and 69.3 (a mean increase from baseline of 18.9) in the aflibercept group (model-based estimate of between-group difference, -0.14; 97.5% CI, -3.07 to ∞; P = .001 for noninferiority), meeting criteria for noninferiority. Ocular adverse events in the aflibercept group included 4 participants with intraocular pressure (IOP) more than 10 mm Hg greater than baseline; ocular adverse events in the bevacizumab group included 1 participant with endophthalmitis (culture negative), 9 with IOP more than 10 mm Hg greater than baseline, 2 with IOP higher than 35 mm Hg, and 1 with angle-closure glaucoma not attributed to the study drug or procedure. CONCLUSIONS AND RELEVANCE: Among patients with macular edema due to central retinal or hemiretinal vein occlusion, intravitreal bevacizumab was noninferior to aflibercept with respect to visual acuity after 6 months of treatment.

Ocular nutritional supplements: are their ingredients and manufacturers' claims evidence-based?

PURPOSE: To compare ingredients contained in top-selling brands of ocular nutritional supplements with the Age-Related Eye Disease Study (AREDS) and AREDS2 formulae and investigate the validity of claims made by manufacturers of leading brands of ocular nutritional supplements. DESIGN: Descriptive. PARTICIPANTS: None. METHODS: We examined the 5 top-selling brands of ocular nutritional supplements in the United States according to dollar sales tracked by SymphonyIRI (Waltham, MA) from June 2011 to June 2012. We reviewed the ingredients and manufacturer claims of 11 ocular nutritional supplements on the companies' consumer information websites; the ingredients were compared with those contained in the AREDS and AREDS2 formulae. MAIN OUTCOME MEASURES: Proportion of ocular nutritional supplements that contained the same ingredients, in the same doses, as the AREDS or AREDS2 formula; proportion of nutritional supplements with unsubstantiated claims made by the manufacturer. RESULTS: All of the ocular nutritional supplements contained the ingredients from the AREDS or AREDS2 formula; 36% (4/11) of the supplements contained equivalent doses of AREDS or AREDS2 ingredients; 55% (6/11) included some information about the AREDS on their consumer information websites. Product descriptions from 4 of the 11 supplements (36%) stated that the supplements were important to maintain general eye health; none of these supplements duplicated the AREDS or AREDS2 formula. All the individual supplements claimed to "support," "protect," "help," or "promote" vision and eye health, but none specified that there is no proven benefit in using nutritional supplements for primary prevention of eye disease. CONCLUSIONS: The majority of top-selling ocular nutritional supplements did not contain the identical ingredient dosages of the AREDS or AREDS2 formula and had product description claims that lacked level 1 evidence, underscoring the importance of ophthalmologists educating their patients on the evidence-based role of nutritional supplements in the management of eye health.

Neuroprotective effects of intravitreal triamcinolone acetonide and dexamethasone implant in rabbit retinas after pars plana vitrectomy and silicone oil injection.

PURPOSE: To investigate potential retinal neuroprotective effects of intravitreal triamcinolone acetonide and dexamethasone implant in rabbits after pars plana vitrectomy and intravitreal silicone oil injection. METHODS: The right eyes of 84 rabbits, divided into 3 groups of 28 rabbits each, underwent standard 3-port pars plana vitrectomy with silicone oil (SO group), silicone oil and intravitreal dexamethasone implant (SO/DEX group), or silicone oil and triamcinolone acetonide (SO/TA group). The retina from the left eye of each rabbit served as a control. The animals were killed at 4 weeks after surgery. Qualitative and quantitative histopathologic analyses were performed 4 weeks after surgery, and investigation for apoptosis was performed using the Tunel assay. RESULTS: Intravitreal triamcinolone acetonide and dexamethasone implant were associated with increased retinal neuronal survival, primarily in the outer nuclear layer, inner nuclear layer, and ganglion cell layer. In the SO group, the cell density in eyes that underwent PPV/SO was 31% lower in the outer nuclear layer, 33% lower in the inner nuclear layer, and 45% lower in the ganglion cell layer compared to control eyes (p < 0.05 for all PPV/SO versus control comparisons). Compared to eyes that underwent PPV/SO, the cell density in eyes treated with triamcinolone was 27% higher in the outer nuclear layer, 66% higher in the inner nuclear layer, and 100% higher in the ganglion cell layer (p < 0.05 for all triamcinolone versus PPV/SO comparisons). Compared to eyes that underwent PPV/SO, the cell density in eyes treated with dexamethasone was 46% higher in the outer nuclear layer, 62% higher in the inner nuclear layer, and 77% higher in the ganglion cell layer (p < 0.05 for all dexamethasone versus PPV/SO comparisons). Analyses using the Tunnel assay demonstrated apoptotic bodies in all eyes in the SO group, compared with none of the eyes in the SO/TA and SO/DEX groups. The presence of cell nuclei stained with 49,6-diamidino-2-phenylindole (DAPI) was demonstrated in all groups. CONCLUSION: In this experimental model of neuroprotection, increased retinal neuronal survival was seen in the steroid-treated groups compared with the controls.

Emerging drugs for diabetic macular edema.

INTRODUCTION: Diabetic macular edema (DME) is the most common cause of visual impairment due to diabetic retinopathy. The treatment of DME has recently undergone a paradigm shift. Traditionally, photocoagulation was standard treatment, but pharmacologic therapies are becoming increasingly used for this purpose. All currently available drug therapies for DME are either anti-VEGF agents or corticosteroids. AREAS COVERED: The pathogenesis of DME involves angiogenesis, inflammation and oxidative stress. The scientific rationale to treat DME through the pharmacologic blockade of VEGF and other pro-angiogenic factors is discussed. The fluocinolone insert is approved for the treatment of DME in several European countries, but not in the US at this time. Some medications that are already approved for other retinal diseases, most prominently aflibercept and the dexamethasone delivery system, have recently obtained approval for DME in the US. Other compounds are being studied in earlier-phase clinical trials. EXPERT OPINION: Pharmacologic treatment of DME will likely become increasingly used, especially for patients with edema involving the fovea. At this time, the two main classes of medication for treatment of DME are anti-VEGF agents and corticosteroids. As we continue to collect clinical trials data, the precise role of individual agents, and the continuing role for photocoagulation, will become more clear.