Progression or Aging? A Deep Learning Approach for Distinguishing Glaucoma Progression from Age-Related Changes in OCT scans.

PURPOSE: To develop deep learning (DL) algorithm to detect glaucoma progression using optical coherence tomography (OCT) images, in the absence of a reference standard. DESIGN: Retrospective cohort study. METHODS: Glaucomatous and healthy eyes with ≥5 reliable peripapillary OCT (Spectralis, Heidelberg Engineering) circle scans were included. A weakly supervised time-series learning model, called Noise Positive-Unlabeled (Noise-PU) DL was developed to classify whether sequences of OCT B-scans showed glaucoma progression. The model used two learning schemes, one to identify age-related changes by differentiating test sequences from glaucoma vs. healthy eyes, and the other to identify test-retest variability based on scrambled OCTs of glaucoma eyes. Both models' bases were convolutional neural networks (CNN) and long short-term memory (LSTM) networks which were combined to form a CNN-LSTM model. Model features were combined and jointly trained to identify glaucoma progression, accounting for age-related loss. The DL model's outcomes were compared with ordinary least squares (OLS) regression of retinal nerve fiber layer (RNFL) thickness over time, matched for specificity. The hit ratio was used as a proxy for sensitivity. RESULTS: 8,785 follow-up sequences of 5 consecutive OCT tests from 3253 eyes (1859 subjects) were included in the study. Mean follow-up time was 3.5±1.6 years. In the test sample, the hit ratios of the DL and OLS methods were 0.498 (95%CI:0.470-0.526) and 0.284 (95%CI:0.258-0.309) respectively (P<0.001) when the specificities were equalized to 95%. CONCLUSION: A DL model was able to identify longitudinal glaucomatous structural changes in OCT B-scans using a surrogate reference standard for progression.

Social history and glaucoma progression: the effect of body mass index, tobacco and alcohol consumption on the rates of structural change in patients with glaucoma.

BACKGROUND/AIMS: Although obesity, tobacco and alcohol consumption were linked to the progression of numerous chronic diseases, an association of these social history aspects with glaucoma progression is not yet determined. This study aims to investigate the effect of body mass index (BMI) and history of tobacco and alcohol use on the rates of retinal nerve fibre layer (RNFL) change over time in glaucoma patients. METHODS: 2839 eyes of 1584 patients with glaucoma from the Duke Ophthalmic Registry were included. Patients had at least two spectral-domain optical coherency tomography (SD-OCT) tests over a minimum 6-month follow-up. Self-reported history of alcohol and tobacco consumption was extracted from electronic health records and mean BMI was calculated. Univariable and multivariable linear mixed models were used to determine the effect of each parameter on RNFL change over time. RESULTS: Mean follow-up time was 4.7±2.1 years, with 5.1±2.2 SD-OCT tests per eye. 43% and 54% of eyes had tobacco or alcohol consumption history, respectively, and 34% were classified as obese. Higher BMI had a protective effect on glaucoma progression (0.014 µm/year slower per each 1 kg/m2 higher; p=0.011). Tobacco and alcohol consumption were not significantly associated with RNFL change rates (p=0.473 and p=0.471, respectively). Underweight subjects presented significantly faster rates of structural loss (-0.768 µm/year; p=0.002) compared with normal weight. CONCLUSIONS: In a large clinical population with glaucoma, habits of tobacco and alcohol consumption showed no significant effect on the rates of RNFL change. Higher BMI was significantly associated with slower rates of RNFL loss.

Visual Field Outcomes in the Primary Tube Versus Trabeculectomy Study.

PURPOSE: To describe visual field outcomes in the Primary Tube Versus Trabeculectomy (PTVT) Study. DESIGN: Cohort analysis of a prospective multicenter randomized clinical trial. SUBJECTS: A total of 155 eyes from 155 subjects randomly assigned to treatment with tube shunt surgery (n=84) or trabeculectomy with mitomycin C (n=71). METHODS: The PTVT Study was a multicenter randomized clinical trial comparing the safety and efficacy of trabeculectomy and tube shunt surgery in eyes without prior intraocular surgery. Subjects underwent standard automated perimetry (SAP) at baseline and annually for five years. SAP tests were deemed reliable if the false positive rate was ≤15%. Tests were excluded if visual acuity was ≤20/400 or loss of ≥2 Snellen lines from baseline were attributed to a non-glaucomatous etiology. Linear mixed-effects models were used to compare rates of change in SAP mean deviation (MD) between the two treatment groups. Intraocular pressure (IOP) control was assessed by percentage of visits with IOP <18 mmHg and mean IOP. OUTCOME MEASURES: Rate of change in SAP MD during follow-up. RESULTS: A total of 730 SAP tests were evaluated, with an average of 4.7 tests per eye. The average SAP MD at baseline was -12.8±8.3 dB in the tube group and -12.0±8.4 dB in the trabeculectomy group (p=0.57). The mean rate of change in SAP MD was -0.32±0.39 dB/year in the trabeculectomy group and -0.47±0.43 dB/year in the tube group (p=0.23). Eyes with mean IOP 14-17.5 mmHg had significantly faster rates of SAP MD loss compared to eyes with mean IOP <14 mmHg (-0.59±0.13 vs. -0.27±0.08 dB/year, p=0.012) and eyes with only 50-75% of visits with IOP <18 mmHg had faster rates than those with 100% of visits with IOP <18 mmHg (-0.90±0.16 vs. -0.29±0.08 dB/year, p<0.001). Multivariable analysis identified older age and worse IOP control as risk factors for faster progression in both treatment groups. CONCLUSIONS: No statistically significant difference in mean rates of visual field change was observed between trabeculectomy and tube shunt surgery in the PTVT Study. Worse IOP control was significantly associated with faster rates of SAP MD loss during follow-up. Older patients were also at risk for faster progression.

Socioeconomic Disparities in Glaucoma Severity at Initial Diagnosis: A Nationwide Electronic Health Record Cohort Analysis.

PURPOSE: To assess disparities in initial disease severity among open-angle glaucoma (OAG) patients. DESIGN: Cross-sectional study. METHODS: In this analysis of Epic Cosmos, an aggregated electronic health record dataset encompassing >213 million patients, OAG patients examined in ophthalmology or optometry clinics between January 1, 2013, and June 1, 2023, were evaluated. OAG severity at presentation was classified as mild, moderate, or severe using International Classification of Disease-10 codes. Demographics, social vulnerability index (SVI) scores, and rural-urban commuting area codes were evaluated as predictors of disease stage using ordinal logistic regression. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated. RESULTS: Of 245,669 patients, 38.1% had mild, 32.5% moderate, and 29.3% severe disease at presentation. In multivariable analyses, significant determinants of worse severity included older age (OR: 1.23 per decade, 95% CI: 1.22-1.23), male sex (OR: 1.37, 95% CI: 1.35-1.39), Black race (OR: 1.61, 95% CI: 1.58-1.65), Hispanic ethnicity (OR: 1.15, 95% CI: 1.11-1.18), non-commercial insurance or uninsured status (OR: 2.53, 95% CI: 2.33-2.74), secondary OAGs (eg, pseudoexfoliative glaucoma - OR: 1.65, 95% CI: 1.58-1.72), and higher socioeconomic SVI scores (OR: 1.25 for highest versus lowest quartile, 95% CI: 1.22-1.28). Black and Hispanic patients were diagnosed at younger ages compared to White patients (mean ages: 67.8 ± 12.3 and 68.1 ± 12.8 vs 73.3 ± 11.8 years respectively, P < .001). CONCLUSIONS: Worse OAG at presentation was associated with older age, male sex, Black race, Hispanic ethnicity, non-commercial insurance or uninsured status, secondary OAGs, and greater socioeconomic vulnerability in this nationwide cohort. These findings can help tailor screening programs towards vulnerable populations.

Performance of Linear Mixed Models in Estimating Structural Rates of Glaucoma Progression Using Varied Random Effect Distributions.

Purpose: To compare how linear mixed models (LMMs) using Gaussian, Student t, and log-gamma (LG) random effect distributions estimate rates of structural loss in a glaucomatous population using OCT and to compare model performance to ordinary least squares (OLS) regression. Design: Retrospective cohort study. Subjects: Patients in the Bascom Palmer Glaucoma Repository (BPGR). Methods: Eyes with ≥ 5 reliable peripapillary retinal nerve fiber layer (RNFL) OCT tests over ≥ 2 years were identified from the BPGR. Retinal nerve fiber layer thickness values from each reliable test (signal strength ≥ 7/10) and associated time points were collected. Data were modeled using OLS regression as well as LMMs using different random effect distributions. Predictive modeling involved constructing LMMs with (n - 1) tests to predict the RNFL thickness of subsequent tests. A total of 1200 simulated eyes of different baseline RNFL thickness values and progression rates were developed to evaluate the likelihood of declared progression and predicted rates. Main Outcome Measures: Model fit assessed by Watanabe-Akaike information criterion (WAIC) and mean absolute error (MAE) when predicting future RNFL thickness values; log-rank test and median time to progression with simulated eyes. Results: A total of 35 862 OCT scans from 5766 eyes of 3491 subjects were included. The mean follow-up period was 7.0 ± 2.3 years, with an average of 6.2 ± 1.4 tests per eye. The Student t model produced the lowest WAIC. In predictive models, all LMMs demonstrated a significant reduction in MAE when estimating future RNFL thickness values compared with OLS (P < 0.001). Gaussian and Student t models were similar and significantly better than the LG model in estimating future RNFL thickness values (P < 0.001). Simulated eyes confirmed LMM performance in declaring progression sooner than OLS regression among moderate and fast progressors (P < 0.01). Conclusions: LMMs outperformed conventional approaches for estimating rates of OCT RNFL thickness loss in a glaucomatous population. The Student t model provides the best model fit for estimating rates of change in RNFL thickness, although the use of the Gaussian or Student t distribution in models led to similar improvements in accurately estimating RNFL loss. Financial Disclosures: Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Short-term Detection of Fast Progressors in Glaucoma: The Fast Progression Assessment through Clustered Evaluation (Fast-PACE) Study.

PURPOSE: To evaluate the performance of an intensive, clustered testing approach in identifying eyes with rapid glaucoma progression over 6 months in the Fast Progression Assessment through Clustered Evaluation (Fast-PACE) Study. DESIGN: Prospective cohort study. PARTICIPANTS: A total of 125 eyes from 65 primary open-angle glaucoma (POAG) subjects. METHODS: Subjects underwent 2 sets of 5 weekly visits (clusters) separated by an average of 6 months and then were followed with single visits every 6 months for an overall mean follow-up of 25 months (mean of 17 tests). Each visit consisted of testing with standard automated perimetry (SAP) 24-2 and 10-2, and spectral-domain OCT (SD-OCT). Progression was assessed using trend analyses of SAP mean deviation (MD) and retinal nerve fiber layer (RNFL) thickness. Generalized estimating equations were applied to adjust for correlations between eyes for confidence interval (CI) estimation and hypothesis testing. MAIN OUTCOME MEASURES: Diagnostic accuracy of the 6-month clustering period to identify progression detected during the overall follow-up. RESULTS: A total of 19 of 125 eyes (15%, CI, 9%-24%) progressed based on SAP 24-2 MD over the 6-month clustering period. A total of 14 eyes (11%, CI, 6%-20%) progressed on SAP 10-2 MD, and 16 eyes (13%, CI, 8%-21%) progressed by RNFL thickness, with 30 of 125 eyes (24%, CI, 16%-34%) progressing by function, structure, or both. Of the 35 eyes progressing during the overall follow-up, 25 had progressed during the 6-month clustering period, for a sensitivity of 71% (CI, 53%-85%). Of the 90 eyes that did not progress during the overall follow-up, 85 also did not progress during the 6-month period, for a specificity of 94% (CI, 88%-98%). Of the 14 eyes considered fast progressors by SAP 24-2, SAP 10-2, or SD-OCT during the overall follow-up, 13 were identified as progressing during the 6-month cluster period, for a sensitivity of 93% (CI, 66%-100%) for identifying fast progression with a specificity of 85% (CI, 77%-90%). CONCLUSIONS: Clustered testing in the Fast-PACE Study detected fast-progressing glaucoma eyes over 6 months. The methodology could be applied in clinical trials investigating interventions to slow glaucoma progression and may be of value for short-term assessment of high-risk subjects. FINANCIAL DISCLOSURE(S): Proprietary or commercial disclosure may be found after the references in the Footnotes and Disclosures at the end of this article.

Single Administration of Bimatoprost Implant: Effects on 24-Hour Intraocular Pressure and 1-Year Outcomes.

PURPOSE: To evaluate the effects of a single bimatoprost implant administration on 24-hour intraocular pressure (IOP) lowering at 8 weeks, and 1-year IOP-lowering efficacy and safety outcomes. DESIGN: Multicenter, open-label, 12-month, phase 3b study (NCT04285580). PARTICIPANTS: Adults with open-angle glaucoma or ocular hypertension. METHODS: Participants (n = 31) received 10-µg bimatoprost implant in the study eye on day 1; IOP (sitting and/or supine) was measured with pneumatonometry every 2 hours throughout a 24-hour period at baseline and week 8. IOP was measured by Goldmann applanation tonometry (GAT) at hour 0 (8 am ± 1 hour) at baseline, weeks 8 and 16, and months 6, 9, and 12. MAIN OUTCOME MEASURES: The primary endpoint was the week-8 hour-matched change from baseline in habitual position IOP over 24 hours assessed with pneumatonometry. Hour 0 IOP change from baseline measured with GAT in study eyes that received no additional (rescue) IOP-lowering treatment, treatment-emergent adverse events (TEAEs), and central corneal endothelial cell density (CECD) were evaluated through 12 months. RESULTS: The mean (standard deviation [SD]) baseline IOP at hour 0 was 24.2 (2.70) mmHg and 25.3 (7.15) mmHg by GAT and pneumatonometry, respectively. Pneumatonometer measurements of IOP taken over 24 hours at week 8 with the participant in habitual position (sitting from 8 am to 10 pm, supine from 12 am to 6 am) showed consistent IOP lowering through the day and night and reduced fluctuation in IOP. The range in IOP measurements over 24 hours was reduced from baseline by a mean (SD) of -1.6 (2.98) mmHg. All 31 bimatoprost implant-treated participants completed the 12-month study; 23 (74%) required no rescue IOP-lowering treatment. The mean (SD) IOP reduction from baseline at month 12 in nonrescued eyes was -4.3 (3.35) mmHg. The most common TEAE was conjunctival hyperemia (incidence 35.5%, 11/31). No implant-treated eye had a ≥ 15% loss in CECD from baseline. CONCLUSIONS: A single intracameral administration of the bimatoprost implant lowered IOP in the habitual position consistently throughout the day and night at week 8. The majority of participants continued to have reduced IOP for 1 year without additional therapy. The 1-year safety profile was favorable. FINANCIAL DISCLOSURE(S): Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Hazard Detection During Simulated Driving in Glaucoma Patients.

PRCIS: In this cross-sectional study, glaucoma patients showed slower reaction times (RTs) to hazardous situations when compared with control subjects during simulated driving. Worse RTs were associated with a greater magnitude of visual field loss. PURPOSE: The purpose of this study was to evaluate the impact of different hazardous traffic conditions on driving performance in glaucoma patients using a high-fidelity driving simulator. METHODS: The cross-sectional study was performed with 52 glaucoma patients and 15 control subjects. A series of hazard scenarios were presented, such as pedestrians crossing the street unexpectedly or vehicles suddenly pulling into the driver's lane. RTs in seconds (s) from first the evidence of a hazard to the time it took the driver to take the foot off the gas pedal ("Gas Off") and the time it took to depress the brake pedal ("Brake On") were compared between groups. RESULTS: Overall, mean RTs were statistically significantly slower in glaucoma patients (3.39±3.88 s) compared with controls (2.39±1.99 s; P =0.005) for the "Brake On" task but not for the "Gas Off" task (2.74±3.42 vs. 2.13±1.91 s, respectively; P =0.120). For subjects with glaucoma, multivariable models adjusted for age, gender, race, and visual acuity demonstrated significantly slower RTs for worse values of binocular mean sensitivity for both "Gas Off" and "Brake On" tasks (1.12 and 1.14 s slower per 10 dB worse; P =0.009 and P <0.001, respectively). Subjects with glaucoma took significantly longer times to brake for smaller (low saliency) hazards compared with larger (high saliency) hazards ( P =0.027). CONCLUSIONS: RTs in response to hazardous driving situations were slower for glaucoma patients compared with controls. Individualized assessment of driving fitness using hazardous scenarios in driving simulators could be helpful in providing an assessment of driving risk in glaucoma patients.

Improved Prediction of Perimetric Loss in Glaucomatous Eyes Using Latent Class Mixed Modeling.

PURPOSE: To evaluate whether the identification of distinct classes within a population of glaucoma patients improves estimates of future perimetric loss. DESIGN: Longitudinal cohort study. PARTICIPANTS: A total of 6558 eyes of 3981 subjects from the Duke Ophthalmic Registry with ≥ 5 reliable standard automated perimetry (SAP) tests and ≥ 2 years of follow-up. METHODS: Standard automated perimetry mean deviation (MD) values were extracted with associated timepoints. Latent class mixed models (LCMMs) were used to identify distinct subgroups (classes) of eyes according to rates of perimetric change over time. Rates for individual eyes were then estimated by considering both individual eye data and the most probable class membership for that eye. Data were split into training (80%) and test sets (20%), and test set mean squared prediction errors (MSPEs) were estimated using LCMM and ordinary least squares (OLS) regression. MAIN OUTCOME MEASURES: Rates of change in SAP MD in each class and MSPE. RESULTS: The dataset contained 52 900 SAP tests with an average of 8.1 ± 3.7 tests per eye. The best-fitting LCMM contained 5 classes with rates of -0.06, -0.21, -0.87, -2.15, and +1.28dB/year (80.0%, 10.2%, 7.5%, 1.3%, and 1.0% of the population, respectively) labeled as slow, moderate, fast, catastrophic progressors, and "improvers" respectively. Fast and catastrophic progressors were older (64.1 ± 13.7 and 63.5 ± 16.9 vs. 57.8 ± 15.8, P < 0.001) and had generally mild-moderate disease at baseline (65.7% and 71% vs. 52%, P < 0.001) than slow progressors. The MSPE was significantly lower for LCMM compared to OLS, regardless of the number of tests used to obtain the rate of change (5.1 ± 0.6 vs. 60.2 ± 37.9, 4.9 ± 0.5 vs. 13.4 ± 3.2, 5.6 ± 0.8 vs. 8.1 ± 1.1, 3.4 ± 0.3 vs. 5.5 ± 1.1 when predicting the fourth, fifth, sixth, and seventh visual fields (VFs) respectively; P < 0.001 for all comparisons). MSPE of fast and catastrophic progressors was significantly lower with LCMM versus OLS (17.7 ± 6.9 vs. 48.1 ± 19.7, 27.1 ± 8.4 vs. 81.3 ± 27.1, 49.0 ± 14.7 vs. 183.9 ± 55.2, 46.6 ± 16.0 vs. 232.4 ± 78.0 when predicting the fourth, fifth, sixth, and seventh VFs respectively; P < 0.001 for all comparisons). CONCLUSIONS: Latent class mixed model successfully identified distinct classes of progressors within a large glaucoma population that seemed to reflect subgroups observed in clinical practice. Latent class mixed models were superior to OLS regression in predicting future VF observations. FINANCIAL DISCLOSURE(S): Proprietary or commercial disclosuremay be found after the references.

Intraocular Pressure and Rates of Macular Thinning in Glaucoma.

PURPOSE: To evaluate the effect of intraocular pressure (IOP) on the rates of macular thickness (ganglion cell layer [GCL] and ganglion cell-inner plexiform layer [GCIPL]) change over time measured by spectral-domain (SD) OCT. DESIGN: Retrospective cohort study. PARTICIPANTS: Overall, 451 eyes of 256 patients with primary open-angle glaucoma. METHODS: Data were extracted from the Duke Ophthalmic Registry, a database of electronic medical records of patients observed under routine clinical care at the Duke Eye Center, and satellite clinics. All records from patients with a minimum of 6 months of follow-up and at least 2 good-quality Spectralis SD-OCT macula scans were included. Linear mixed models were used to investigate the relationship between average IOP during follow-up and rates of GCL and GCIPL thickness change over time. MAIN OUTCOME MEASURES: The effect of IOP on the rates of GCL and GCIPL thickness loss measured by SD-OCT. RESULTS: Eyes had a mean follow-up of 1.8 ± 1.3 years, ranging from 0.5 to 10.2 years. The average rate of change for GCL thickness was -0.220 µm/year (95% confidence interval [CI], -0.268 to -0.172 µm/year) and for GCIPL thickness was -0.231 µm/year (95% CI, -0.302 to -0.160 µm/year). Each 1-mmHg higher mean IOP during follow-up was associated with an additional loss of -0.021 µm/year of GCL thickness (P = 0.001) and -0.032 µm/year of GCIPL thickness (P = 0.001) after adjusting for potentially confounding factors, such as baseline age, disease severity, sex, race, central corneal thickness, and follow-up time. CONCLUSIONS: Higher IOP was significantly associated with faster rates of GCL and GCIPL loss over time measured by SD-OCT, even during relatively short follow-up times. These findings support the use of SD-OCT GCL and GCIPL thickness measurements as structural biomarkers for the evaluation of the efficacy of IOP-lowering therapies in slowing down the progression of glaucoma. FINANCIAL DISCLOSURE(S): Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Systematic User-centered Design of a Prototype Clinical Decision Support System for Glaucoma.

Purpose: To rigorously develop a prototype clinical decision support (CDS) system to help clinicians determine the appropriate timing for follow-up visual field testing for patients with glaucoma and to identify themes regarding the context of use for glaucoma CDS systems, design requirements, and design solutions to meet these requirements. Design: Semistructured qualitative interviews and iterative design cycles. Participants: Clinicians who care for patients with glaucoma, purposefully sampled to ensure a representation of a range of clinical specialties (glaucoma specialist, general ophthalmologist, optometrist) and years in clinical practice. Methods: Using the established User-Centered Design Process framework, we conducted semistructured interviews with 5 clinicians that addressed the context of use and design requirements for a glaucoma CDS system. We analyzed the interviews using inductive thematic analysis and grounded theory to generate themes regarding the context of use and design requirements. We created design solutions to address these requirements and used iterative design cycles with the clinicians to refine the CDS prototype. Main Outcome Measures: Themes regarding decision support for determining the timing of visual field testing for patients with glaucoma, CDS design requirements, and CDS design features. Results: We identified 9 themes that addressed the context of use for the CDS system, 9 design requirements for the prototype CDS system, and 9 design features intended to address these design requirements. Key design requirements included the preservation of clinician autonomy, incorporation of currently used heuristics, compilation of data, and increasing and communicating the level of certainty regarding the decision. After completing 3 iterative design cycles using this preliminary CDS system design solution, the design was satisfactory to the clinicians and was accepted as our prototype glaucoma CDS system. Conclusions: We used a systematic design process based on the established User-Centered Design Process to rigorously develop a prototype glaucoma CDS system, which will be used as a starting point for a future, large-scale iterative refinement and implementation process. Clinicians who care for patients with glaucoma need CDS systems that preserve clinician autonomy, compile and present data, incorporate currently used heuristics, and increase and communicate the level of certainty regarding the decision. Financial Disclosures: Proprietary or commercial disclosure may be found after the references.

The Definition of Glaucomatous Optic Neuropathy in Artificial Intelligence Research and Clinical Applications.

OBJECTIVE: Although artificial intelligence (AI) models may offer innovative and powerful ways to use the wealth of data generated by diagnostic tools, there are important challenges related to their development and validation. Most notable is the lack of a perfect reference standard for glaucomatous optic neuropathy (GON). Because AI models are trained to predict presence of glaucoma or its progression, they generally rely on a reference standard that is used to train the model and assess its validity. If an improper reference standard is used, the model may be trained to detect or predict something that has little or no clinical value. This article summarizes the issues and discussions related to the definition of GON in AI applications as presented by the Glaucoma Workgroup from the Collaborative Community for Ophthalmic Imaging (CCOI) US Food and Drug Administration Virtual Workshop, on September 3 and 4, 2020, and on January 28, 2022. DESIGN: Review and conference proceedings. SUBJECTS: No human or animal subjects or data therefrom were used in the production of this article. METHODS: A summary of the Workshop was produced with input and approval from all participants. MAIN OUTCOME MEASURES: Consensus position of the CCOI Workgroup on the challenges in defining GON and possible solutions. RESULTS: The Workshop reviewed existing challenges that arise from the use of subjective definitions of GON and highlighted the need for a more objective approach to characterize GON that could facilitate replication and comparability of AI studies and allow for better clinical validation of proposed AI tools. Different tests and combination of parameters for defining a reference standard for GON have been proposed. Different reference standards may need to be considered depending on the scenario in which the AI models are going to be applied, such as community-based or opportunistic screening versus detection or monitoring of glaucoma in tertiary care. CONCLUSIONS: The development and validation of new AI-based diagnostic tests should be based on rigorous methodology with clear determination of how the reference standards for glaucomatous damage are constructed and the settings where the tests are going to be applied. FINANCIAL DISCLOSURE(S): Proprietary or commercial disclosure may be found after the references.

Identifying Risk Factors for Blindness From Glaucoma at First Presentation to a Tertiary Clinic.

PURPOSE: Glaucoma is the leading cause of irreversible blindness, a crippling disability resulting in higher risks of chronic health conditions. To better understand disparities in blindness risk, we identified risk factors of blindness on first presentation to a glaucoma clinic using a large clinical database. DESIGN: Retrospective cross-sectional study. METHODS: We used electronic health records of glaucoma patients from the Duke Ophthalmic Registry. International Classification of Diseases codes were used to identify glaucoma and exclude concurrent diseases. Blindness classification was based on the definition of legal blindness. Risk factors included gender, race, marital status, age, intraocular pressure, diabetes history, income level, and education. Odds ratios (ORs) and 95% CIs were calculated for risk factors using univariable and multivariable logistic regression. RESULTS: Our cohort consisted of 3753 patients, with 192 (5%) blind on first presentation. In univariable models, African American / Black race (OR 2.48, 95% CI 1.83-3.36), single marital status (1.74, 95% CI 1.25-2.44), prior diabetes diagnosis (2.23, 95% CI 1.52-3.27), and higher intraocular pressure (1.29 per 1 SD higher, 95% CI 1.13-1.46) were associated with increased risk of presenting blind, whereas higher annual income (0.75, 95% CI 0.65-0.86) and education (0.77, 95% CI 0.69-0.85) were associated with lower risk. These associations remained significant and in the same direction in a multivariable model apart from income, which became insignificant. CONCLUSIONS: Using a large real-world clinical database, we identified risk factors associated with presentation with blindness among glaucoma patients. Our results highlight disparities in health care outcomes and indicate the importance of targeted education to reduce disparities in blindness.

Rates of Visual Field Change in Eyes With Optic Disc Drusen.

BACKGROUND: Optic disc drusen (ODD) are calcified deposits in the prelaminar portion of the optic nerve head. Although often asymptomatic, these deposits can cause progressive visual field defects and vision loss. The purpose of this study was to evaluate rates of functional loss in eyes with ODD and to investigate risk factors associated with rates of visual field progression. METHODS: This was a retrospective cohort study including 65 eyes of 43 patients with ODD from the Duke Ophthalmic Registry. All eyes had at least 12 months of follow-up and at least 3 reliable standard automated perimetry (SAP) tests. Linear mixed models were used to estimate rates of SAP mean deviation (MD) loss over time. Univariable and multivariable models were used to assess the effect of clinical variables and intraocular pressure (IOP) on rates of change. RESULTS: Subjects were followed for an average of 7.6 ± 5.3 years. The mean rate of SAP MD change was -0.23 ± 0.26 dB/year, ranging from -1.19 to 0.13 dB/year. Fifty-seven eyes (87.7%) had slow progression (slower than -0.5 dB/year), 6 eyes (9.2%) had moderate progression (between -0.5 dB/year and -1 dB/year), and 2 eyes (3.1%) had fast progression (faster than -1 dB/year). In multivariable models, older age and worse SAP MD at baseline were significantly associated with faster rates of change. Mean IOP was not associated with faster rates of MD change in both univariable and multivariable analyses. CONCLUSIONS: Most eyes with ODD had slow rates of visual field loss over time. Age and baseline severity were significantly associated with faster rates of visual field loss.

Endpoints for clinical trials in ophthalmology.

With the identification of novel targets, the number of interventional clinical trials in ophthalmology has increased. Visual acuity has for a long time been considered the gold standard endpoint for clinical trials, but in the recent years it became evident that other endpoints are required for many indications including geographic atrophy and inherited retinal disease. In glaucoma the currently available drugs were approved based on their IOP lowering capacity. Some recent findings do, however, indicate that at the same level of IOP reduction, not all drugs have the same effect on visual field progression. For neuroprotection trials in glaucoma, novel surrogate endpoints are required, which may either include functional or structural parameters or a combination of both. A number of potential surrogate endpoints for ophthalmology clinical trials have been identified, but their validation is complicated and requires solid scientific evidence. In this article we summarize candidates for clinical endpoints in ophthalmology with a focus on retinal disease and glaucoma. Functional and structural biomarkers, as well as quality of life measures are discussed, and their potential to serve as endpoints in pivotal trials is critically evaluated.

Fast Progressors in Glaucoma: Prevalence Based on Global and Central Visual Field Loss.

PURPOSE: To determine the prevalence of fast global and central visual field (VF) progression in individuals with glaucoma under routine care. DESIGN: Observational study. PARTICIPANTS: Six hundred ninety-three eyes of 461 individuals with glaucoma followed up over a median of 4.5 years. METHODS: This study included (1) patients at a private ophthalmology clinic in Melbourne, Australia, and (2) individuals in 2 prospective longitudinal observational studies across 3 sites in the United States. All individuals had a diagnosis of glaucoma and were under routine care, and had performed 5 or more reliable 24-2 VF tests over a 1- to 5-year period. Ordinary least squares regression analyses were used to calculate the rate of global mean deviation (MD) change over time and the rate of the mean total deviation values of the 12 test locations within the central 10° region (MTD10) for each eye. MAIN OUTCOME MEASURES: Prevalence of progression based on the rate of MD and the MTD10 change across various fixed cutoffs and cutoffs based on the estimated normal distribution (from the positive slopes). RESULTS: Based on the MD and the MTD10, 12.5% and 11.7% of the eyes, respectively, exhibited a rate of change that was less than -1.0 dB/year (being a rate that typically is defined as "fast progression" for MD values), and 29.0% of the eyes showed a change of less than -0.5 dB/year on MTD10. Furthermore, 12.7% and 9.1% of the eyes exhibited a rate of change that exceeded the 1% cutoff of the estimated normal distribution MD and the MTD10 values, respectively. CONCLUSIONS: This study found that approximately 1 in 8 eyes with glaucoma receiving routine care showed fast progression based on global MD values (< -1.0 dB/year) and that nearly 1 in 3 eyes showed a < -0.5 dB/year decline centrally. These findings highlight the clinical importance of assessing progressive central VF loss and reinforce the need for new therapies to prevent functional disability in a notable proportion of individuals who continue to exhibit fast progression. FINANCIAL DISCLOSURE(S): Proprietary or commercial disclosure may be found after the references.

Deep Learning-Assisted Detection of Glaucoma Progression in Spectral-Domain OCT.

PURPOSE: To develop and validate a deep learning (DL) model for detection of glaucoma progression using spectral-domain (SD)-OCT measurements of retinal nerve fiber layer (RNFL) thickness. DESIGN: Retrospective cohort study. PARTICIPANTS: A total of 14 034 SD-OCT scans from 816 eyes from 462 individuals. METHODS: A DL convolutional neural network was trained to assess SD-OCT RNFL thickness measurements of 2 visits (a baseline and a follow-up visit) along with time between visits to predict the probability of glaucoma progression. The ground truth was defined by consensus from subjective grading by glaucoma specialists. Diagnostic performance was summarized by the area under the receiver operator characteristic curve (AUC), sensitivity, and specificity, and was compared with conventional trend-based analyses of change. Interval likelihood ratios were calculated to determine the impact of DL model results in changing the post-test probability of progression. MAIN OUTCOME MEASURES: The AUC, sensitivity, and specificity of the DL model. RESULTS: The DL model had an AUC of 0.938 (95% confidence interval [CI], 0.921-0.955), with sensitivity of 87.3% (95% CI, 83.6%-91.6%) and specificity of 86.4% (95% CI, 79.9%-89.6%). When matched for the same specificity, the DL model significantly outperformed trend-based analyses. Likelihood ratios for the DL model were associated with large changes in the probability of progression in the vast majority of SD-OCT tests. CONCLUSIONS: A DL model was able to assess the probability of glaucomatous structural progression from SD-OCT RNFL thickness measurements. The model agreed well with expert judgments and outperformed conventional trend-based analyses of change, while also providing indication of the likely locations of change. FINANCIAL DISCLOSURE(S): Proprietary or commercial disclosure may be found after the references.

Validation of Rates of Mean Deviation Change as Clinically Relevant End Points for Glaucoma Progression.

PURPOSE: To investigate whether rates of standard automated perimetry (SAP) mean deviation (MD) over an initial 2-year follow-up period were predictive of events of visual field progression over an extended follow-up. DESIGN: Longitudinal, prospective, observational study. PARTICIPANTS: Two hundred forty-six eyes of 168 patients with glaucoma followed up every 6 months for up to 5 years. METHODS: Patients were required to have a minimum of 5 reliable SAP tests during the first 2 years of follow-up. Events of progression were evaluated using 2 methods: Guided Progression Analysis (GPA; Carl Zeiss Meditec, Inc) and a United States Food and Drug Administration (FDA)-suggested end point. The date of the first test showing progression after the first 2 years was considered to be the event date. Rates of change in SAP MD were calculated for the first 2 years of follow-up, and joint longitudinal survival models were used to assess the risk of faster initial MD loss for subsequent progression based on each event analysis. MAIN OUTCOME MEASURE: Risk of having an event of progression based on initial rates of SAP MD change. RESULTS: Fifty-six eye (22.8%) showed an event of progression by the GPA and 51 eyes (20.7%) did so by the FDA end point. Each 0.1-dB/year faster rate of SAP MD loss in the first 2 years was associated with a 26% increase in risk of a GPA progression end point developing (R2 = 76%) and 32% risk of an FDA-based end point developing (R2 = 83%). A reduction of 30% in the rate of MD change in the first 2 years was associated with a 20% reduction in the cumulative probability of a progression event developing over 5 years of follow-up. CONCLUSIONS: Rates of SAP MD change for eyes with glaucoma calculated over the initial 2 years of follow-up were strongly predictive of events of progression over subsequent follow-up. Our findings give support for the use of slopes of MD change as suitable end points of progression in clinical trials. FINANCIAL DISCLOSURE(S): Proprietary or commercial disclosure may be found after the references.

Bimatoprost Implant Biodegradation in the Phase 3, Randomized, 20-Month ARTEMIS Studies.

Purpose: To evaluate the time course of biodegradation of an intracameral, biodegradable, sustained-release bimatoprost implant that lowers intraocular pressure without the need for daily eye drops. Methods: In 2 identically designed, randomized, phase 3 clinical trials, adults with open-angle glaucoma or ocular hypertension and open iridocorneal angles inferiorly in the study eye were administered 10- or 15-µg bimatoprost implant (day 1 and weeks 16 and 32) or twice-daily topical timolol 0.5%. Implants were assessed on gonioscopy throughout the studies. Investigators reported whether implants were visible, estimated the size of visible implants relative to their initial size at implantation, and reported the implant location. Data for 10-µg implant placed on day 1 were pooled from both studies for analysis. Results: A total of 372 patients received the 10-µg bimatoprost implant. The degree of implant biodegradation at each follow-up time point was variable among patients. The implant frequently swelled during the initial phase of biodegradation from 6 to 28 weeks. Accelerated biodegradation occurred between 31 and 52 weeks, resulting in 82% of implants absent or ≤25% of initial size by 52 weeks. By month 20, 95% of implants had biodegraded to absent or ≤25% of initial size. The implant was predominantly located inferiorly in the iridocorneal angle. Conclusions: Bimatoprost implant biodegradation in phase 3 studies showed some degree of variability among patients. Clinically significant implant biodegradation was observed in the majority of patients by 12 months. Clinical studies are in progress to further understand implant biodegradation and the ideal timing for implant re-administration. ClinicalTrials.gov NCT02247804; ClinicalTrials.gov NCT02250651.

Association of an Objective Structural and Functional Reference Standard for Glaucoma with Quality of Life Outcomes.

PURPOSE: To compare self-reported quality-of-life (QoL) outcomes of patients diagnosed as normal, glaucoma suspect, and glaucoma based on an objective reference standard for glaucomatous optic neuropathy (GON). DESIGN: Cross-sectional study. PARTICIPANTS: 1884 eyes of 1019 patients were included in the study. METHODS: The data was sourced from the Duke Glaucoma Registry. Eyes were classified according to the presence and topographic correspondence of functional and structural damage, as assessed by parameters from standard automated perimetry (SAP) and spectral-domain OCT (SD-OCT). The objective diagnosis of the worse eye was used to define patient-level diagnosis. To assess QoL in the diagnostic groups, 14 unidimensional vision-related items of the National Eye Institute Visual Functioning Questionnaire (NEI VFQ-25) were used to assess QoL in the diagnostic groups. Association between NEI VFQ-25 Rasch-calibrated scores and diagnostic groups was assessed through multivariable regression that controlled for confounding demographic and socioeconomic variables such as age, sex, race, income, marriage status, insurance status, and highest education level. MAIN OUTCOME MEASURES: NEI VFQ-25 Rasch scores compared with objective criteria diagnosis based on SAP mean deviation (MD) and SD-OCT retinal nerve fiber layer (RNFL) thickness. RESULTS: Overall, eyes classified as normal, glaucoma suspect, and glaucoma had decreasing mean scores in SAP MD (0.2 ± 1.0 dB, -0.9 ± 2.4 dB, -6.2 ± 7.0 dB, respectively; P < 0.001) and SD-OCT RNFL thickness (97.8 ± 9.5 µm, 89.0 ± 13.1 µm, 64.5 ± 12.8 µm, respectively; P < 0.001). The mean Rasch-calibrated NEI VFQ-25 score was significantly different among normal, suspect, and glaucoma groups (82.9 ± 13.0, 78.2 ± 14.8, and 72.6 ± 16.2, respectively; P < 0.001). When adjusted for confounding socioeconomic variables, glaucoma patients had significantly worse QoL than those classified as normal (ß = -6.8 Rasch score units; P < 0.001). CONCLUSION: A glaucoma diagnosis, based on an objective reference standard for GON, was significantly associated with worse Rasch-adjusted scores of QoL. Utilization of such objective criteria may provide clinically relevant metrics with potential to improve comparability of research findings and validation of newly proposed diagnostic tools. FINANCIAL DISCLOSURE(S): Proprietary or commercial disclosure may be found after the references.