Effectiveness and Safety of Trabeculectomy Versus Tube Shunt Implantation for Uveitic Glaucoma: A Systematic Review and Meta-analysis.

PURPOSE: To evaluate the effectiveness and safety of trabeculectomy compared to glaucoma drainage devices (GDDs) in managing uveitic glaucoma (UG). DESIGN: Systematic review METHODS: : We searched seven electronic databases [PubMed, Scopus, Web of Science, ScienceDirect, EMBASE, CENTRAL, and Google Scholar] to compare trabeculectomy with various GDDs in UG. The primary outcome was intraocular pressure (IOP) reduction, and secondary outcomes included postoperative complications. We fitted a random effects model for meta-analysis and assessed the risk of bias using the National Institute of Health quality assessment tool. RESULTS: We included eight studies; 197 eyes underwent trabeculectomy, and 277 eyes had GDDs. The mean age of participants was 48.5 years, with ∼53.5% being male in the trabeculectomy group, and 49.3% in the GDDs group. The meta-analysis revealed no significant difference in IOP reduction between trabeculectomy and GDDs (p=0.48). Subgroup analyses revealed no significant difference in IOP reduction between trabeculectomy and either the Ahmed glaucoma drainage device group (p =0.38) or the Baerveldt glaucoma implant group (p =0.90). GDDs were associated with higher rates of complications such as cystoid macular edema (CME) (15% vs. 4%, p<0.001), need for revision surgery (11% vs. 6%, P =0.04), and uveitic flare (5% vs. 0%, p =0.001). However, trabeculectomy had a higher risk of cataract progression (7% vs. 1%, p<0.001). CONCLUSION: Trabeculectomy and GDDs demonstrated comparable effectiveness in reducing IOP or glaucoma medication reduction in UG. However, there are significant differences in their safety profiles; CME and revisions were higher in GDD, and cataract progression was higher after trabeculectomy.

Comparison of Diagnosis Codes to Clinical Notes in Classifying Patients with Diabetic Retinopathy.

Purpose: Electronic health records (EHRs) contain a vast amount of clinical data. Improved automated classification approaches have the potential to accurately and efficiently identify patient cohorts for research. We evaluated if a rule-based natural language processing (NLP) algorithm using clinical notes performed better for classifying proliferative diabetic retinopathy (PDR) and nonproliferative diabetic retinopathy (NPDR) severity compared with International Classification of Diseases, ninth edition (ICD-9) or 10th edition (ICD-10) codes. Design: Cross-sectional study. Subjects: Deidentified EHR data from an academic medical center identified 2366 patients aged ≥18 years, with diabetes mellitus, diabetic retinopathy (DR), and available clinical notes. Methods: From these 2366 patients, 306 random patients (100 training set, 206 test set) underwent chart review by ophthalmologists to establish the gold standard. International Classification of Diseases codes were extracted from the EHR. The notes algorithm identified positive mention of PDR and NPDR severity from clinical notes. Proliferative diabetic retinopathy and NPDR severity classification by ICD codes and the notes algorithm were compared with the gold standard. The entire DR cohort (N = 2366) was then classified as having presence (or absence) of PDR using ICD codes and the notes algorithm. Main Outcome Measures: Sensitivity, specificity, positive predictive value (PPV), negative predictive value, and F1 score for the notes algorithm compared with ICD codes using a gold standard of chart review. Results: For PDR classification of the test set patients, the notes algorithm performed better than ICD codes for all metrics. Specifically, the notes algorithm had significantly higher sensitivity (90.5% [95% confidence interval 85.7, 94.9] vs. 68.4% [60.4, 75.3]), but similar PPV (98.0% [95.4-100] vs. 94.7% [90.3, 98.3]) respectively. The F1 score was 0.941 [0.910, 0.966] for the notes algorithm compared with 0.794 [0.734, 0.842] for ICD codes. For PDR classification, ICD-10 codes performed better than ICD-9 codes (F1 score 0.836 [0.771, 0.878] vs. 0.596 [0.222, 0.692]). For NPDR severity classification, the notes algorithm performed similarly to ICD codes, but performance was limited by small sample size. Conclusions: The notes algorithm outperformed ICD codes for PDR classification. The findings demonstrate the significant potential of applying a rule-based NLP algorithm to clinical notes to increase the efficiency and accuracy of cohort selection for research. Financial Disclosures: Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Race, Social Determinants of Health, and the Quality of Diabetic Eye Care.

Importance: Besides race, little is known about how other social determinants of health (SDOH) affect quality of diabetic eye care. Objective: To evaluate the association between multiple SDOH and monitoring for diabetic retinopathy (DR) in accordance with clinical practice guidelines (CPGs). Design, Setting, and Participants: This cohort study was conducted in 11 US medical centers and included adult patients (18-75 years old) with diabetes. Patients received care from 2012 to 2023 and had 18 months or more of follow-up. Exposures: Multiple SDOH and associated factors, including ethnicity, urbanicity of residence, health insurance type, and diabetes type. Main Outcomes and Measures: Adjusted odds ratio (aOR) of receiving 1 or more eye-care visits and 1 or more dilated fundus examinations in accordance with CPGs. Results: The study cohort included 37 397 adults with diabetes: 10 157 Black patients and 27 240 White patients. The mean (SD) age was 58 (11) years for Black patients and 59 (11) years for White patients. Of the Black patients, 6422 (63.2%) were female and 3735 (36.8%) male; of the White patients, 13 120 (48.1) were female and 14 120 (51.8) were male. Compared with those of the same race in urban communities, Black patients (aOR, 0.12; 95% CI, 0.04-0.31) and White patients (aOR, 0.75; 95% CI, 0.62-0.91) with diabetes living in rural communities had 88% and 25% lower odds of having eye-care visits, respectively. Sicker Black and White patients, defined by the Charlson Comorbidity Index, had 4% (aOR, 1.04; 95% CI, 1.02-1.06) and 5% (aOR, 1.05, CI 1.04-1.06) higher odds of having an eye-care visit, respectively. Black patients with preexisting DR had 15% lower odds of visits (aOR, 0.85, CI 0.73-0.99) compared with those without preexisting DR while White patients with preexisting DR had 16% higher odds of eye-care visits (aOR, 1.16; 95% CI, 1.05-1.28). White patients with Medicare (aOR, 0.85; 95% CI, 0.80-0.91) and Medicaid (aOR, 0.81; 95% CI, 0.68-0.96) had lower odds of eye-care visits vs patients with commercial health insurance. Hispanic White patients had 15% lower odds of eye-care visits (aOR, 0.85; 95% CI, 0.74-0.98) vs non-Hispanic White patients. White patients with type 1 diabetes had 17% lower odds of eye-care visits (aOR, 0.83; 95% CI, 0.76-0.90) vs those with type 2 diabetes. Among patients who had eye-care visits, those with preexisting DR (Black: aOR, 1.68; 95% CI, 1.11-2.53; White: aOR, 1.51; 95% CI, 1.16-1.96) were more likely to undergo dilated fundus examinations. Conclusions and Relevance: This study found that certain SDOH affected monitoring for DR similarly for Black and White patients with diabetes while others affected them differently. Patients living in rural communities, Black patients with preexisting DR, and Hispanic White patients were not receiving eye care in accordance with CPGs, which may contribute to worse outcomes.

Demographic and Metabolic Risk Factors Associated with Development of Diabetic Macular Edema among Persons with Diabetes Mellitus.

Purpose: Diabetic macular edema (DME), a leading cause of visual impairment, can occur regardless of diabetic retinopathy (DR) stage. Poor metabolic control is hypothesized to contribute to DME development, although large-scale studies have yet to identify such an association. This study aims to determine whether measurable markers of dysmetabolism are associated with DME development in persons with diabetes. Design: Retrospective cohort study. Participants: Using data from the Sight Outcomes Research Collaborative (SOURCE) repository, patients with diabetes mellitus and no preexisting DME were identified and followed over time to see what factors associated with DME development. Methods: Cox proportional hazard modeling was used to assess the relationship between demographic variables, diabetes type, smoking history, baseline DR status, blood pressure (BP), lipid profile, body mass index (BMI), hemoglobin A1C (HbA1C), and new onset of DME. Main Outcome Measures: Adjusted hazard ratio (HR) of developing DME with 95% confidence intervals (CIs). Results: Of 47 509 eligible patients from 10 SOURCE sites (mean age 63 ± 12 years, 58% female sex, 48% White race), 3633 (7.6%) developed DME in the study period. The mean ± standard deviation time to DME was 875 ± 684 days (∼2.4 years) with those with baseline nonproliferative DR (HR 3.67, 95% CI: 3.41-3.95) and proliferative DR (HR 5.19, 95% CI: 4.61-5.85) more likely to develop DME. There was no difference in DME risk between type 1 and type 2 patients; however, Black race was associated with a 40% increase in DME risk (HR 1.40, 95% CI: 1.30-1.51). Every 1 unit increase in HbA1C had a 15% increased risk of DME (HR 1.15, 95% CI: 1.13-1.17), and each 10 mmHg increase in systolic BP was associated with a 6% increased DME risk (HR 1.06, 95% CI: 1.02-1.09). No association was identified between DME development and BMI, triglyceride levels, or high-density lipoprotein levels. Conclusions: These findings suggest that in patients with diabetes modifiable risk factors such as elevated HbA1C and BP confer a higher risk of DME development; however, other modifiable systemic markers of dysmetabolism such as obesity and dyslipidemia did not. Further work is needed to identify the underlying contributions of race in DME. Financial Disclosures: Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Gap Analysis of Glaucoma Examination Concept Representations within Standard SNOMED Clinical Terms.

PURPOSE: Standardization of eye care data is important for clinical interoperability and research . We aimed to address gaps in the representations of glaucoma examination concepts within Systemized Nomenclature of Medicine - Clinical Terms (SNOMED-CT), the preferred terminology of the American Academy of Ophthalmology. DESIGN: Study of data elements. METHODS: Structured eye exam data fields from two electronic health records (EHR) systems (Epic Systems and Medisoft) were compared against existing SNOMED-CT codes for concepts representing glaucoma examination findings3. Glaucoma specialists from multiple institutions were surveyed to identify high-priority gaps in representation, which were discussed among the SNOMED International Eye Care Clinical Reference Group. Proposals for new codes to address the gaps were formulated and submitted for inclusion in SNOMED-CT. MAIN OUTCOME MEASURES: Gaps in SNOMED-CT glaucoma examination concept representations RESULTS: We identified several gaps in SNOMED-CT regarding glaucoma examination concepts. A survey of glaucoma specialists identified high-priority data elements within the categories of tonometry and gonioscopy. For tonometry, there was consensus that we need to define new codes related to maximum intraocular pressure (IOP) and target IOP, and to delineate all methods of measuring IOP. These new codes were proposed and successfully added to SNOMED-CT for future use. Regarding gonioscopy, the current terminology did not include the ability to denote the gonioscopic grading system used (e.g., Shaffer or Spaeth), degree of angle pigmentation, iris configuration (except for plateau iris), and iris approach. There was also no ability to specify eye laterality or angle quadrant for gonioscopic findings. We proposed a framework for representing gonioscopic findings as observable entities in SNOMED-CT. DISCUSSION: There are existing gaps in the standardized representation of findings related to tonometry and gonioscopy within SNOMED-CT. These are important areas for evaluating clinical outcomes and enabling secondary use of EHR data for glaucoma research. This international, multi-institutional collaborative process enabled identification of gaps, prioritization, and development of data standards to address these gaps. CONCLUSION: Addressing these gaps and augmenting SNOMED-CT coverage of glaucoma examination findings could enhance clinical documentation and future research efforts related to glaucoma.

Incidence of Acute Cystoid Macular Edema After Starting a Prostaglandin Analog Compared With Other Classes of Glaucoma Medications.

PURPOSE: There is a longstanding belief that prostaglandin analogs (PGAs) may predispose patients with glaucoma to develop acute cystoid macular edema (CME). However, there is little solid evidence supporting this notion. The purpose of this study is to compare CME incidence rates among patients initiating treatment with different glaucoma medication classes. DESIGN: Database study. PARTICIPANTS: 39948 patients who were newly prescribed glaucoma medications METHODS: Using data from 10 health systems contributing data to the Sight Outcomes Research Collaborative (SOURCE) Ophthalmology Data Repository, we identified all adults with glaucoma who had been newly started on a topical glaucoma medication. Patients with pre-existing documentation of macular edema were excluded. We assessed the incidence of CME among patients with glaucoma who were newly started on PGAs, topical beta blockers (BBs), alpha agonists (AAs), and carbonic anhydrase inhibitors (CAIs). Using multivariable logistic regression, and adjusting for sociodemographic factors, we assessed the odds of developing CME among patients prescribed each of the 4 glaucoma medication classes. We also performed a subset regression analysis including lens status as a co-variate. MAIN OUTCOME MEASURES: Incidence of CME within 3 months of initiating therapy with different topical glaucoma medications. RESULTS: Among the 39,948 patients were newly treated with a topical glaucoma medication, 139 (0.35%) developed CME. The incidence of CME was 0.13%, 0.65%, 0.55%, 1.76% for users of PGAs, BBs, alpha agonists (AAs) and carbonic anhydrase inhibitors (CAIs), respectively. After adjusting for sociodemographic factors, users of topical BBs, AAs and CAIs had substantially higher odds of developing CME compared with PGA users (P<0.001 for all comparisons). The subset analysis also showed higher odds ratio of the non-PGA medication classes in association with CME. CONCLUSIONS: Clinicians should reconsider the notion that PGAs carry a higher risk of CME versus other glaucoma medication classes. If additional studies support the findings of these analyses, clinicians may feel more comfortable prescribing PGAs to patients with glaucoma without fear they will predispose patients to CME.

Assessment of Social Vulnerabilities of Care and Prognosis in Adult Ocular Melanomas in the US.

BACKGROUND: Prior works have studied the impact of social determinants on various cancers but there is limited analysis on eye-orbit cancers. Current literature tends to focus on socioeconomic status and race, with sparse analysis of interdisciplinary contributions. We examined social determinants as measured by the Centers for Disease Control and Prevention (CDC) Social Vulnerability Index (SVI), quantifying eye and orbit melanoma disparities across the United States. METHODS: A retrospective review of 15,157 patients diagnosed with eye-orbit cancers in the Surveillance, Epidemiology, and End Results (SEER) database from 1975 to 2017 was performed, extracting 6139 ocular melanomas. SVI scores were abstracted and matched to SEER patient data, with scores generated by weighted averages per population density of county's census tracts. Primary outcome was months survived, while secondary outcomes were advanced staging, high grading, and primary surgery receipt. RESULTS: With increased total SVI score, indicating more vulnerability, we observed significant decreases of 23.1% in months survival for melanoma histology (p < 0.001) and 19.6-39.7% by primary site. Increasing total SVI showed increased odds of higher grading (odds ratio [OR] 1.20, 95% confidence interval [CI] 1.02-1.43) and decreased odds of surgical intervention (OR 0.94, 95% CI 0.92-0.96). Of the four themes, higher magnitude contributions were observed with socioeconomic status (26.0%) and housing transportation (14.4%), while lesser magnitude contributions were observed with minority language status (13.5%) and household composition (9.0%). CONCLUSIONS: Increasing social vulnerability, as measured by the CDC SVI and its subscores, displayed significant detrimental trends in prognostic and treatment factors for adult eye-orbit melanoma. Subscores quantified which social determinants contributed most to disparities. This lays groundwork for providers to target the highest-impact social determinant for non-clinical factors in patient care.

Prediction Models for Glaucoma in a Multicenter Electronic Health Records Consortium: The Sight Outcomes Research Collaborative.

Purpose: Advances in artificial intelligence have enabled the development of predictive models for glaucoma. However, most work is single-center and uncertainty exists regarding the generalizability of such models. The purpose of this study was to build and evaluate machine learning (ML) approaches to predict glaucoma progression requiring surgery using data from a large multicenter consortium of electronic health records (EHR). Design: Cohort study. Participants: Thirty-six thousand five hundred forty-eight patients with glaucoma, as identified by International Classification of Diseases (ICD) codes from 6 academic eye centers participating in the Sight OUtcomes Research Collaborative (SOURCE). Methods: We developed ML models to predict whether patients with glaucoma would progress to glaucoma surgery in the coming year (identified by Current Procedural Terminology codes) using the following modeling approaches: (1) penalized logistic regression (lasso, ridge, and elastic net); (2) tree-based models (random forest, gradient boosted machines, and XGBoost), and (3) deep learning models. Model input features included demographics, diagnosis codes, medications, and clinical information (intraocular pressure, visual acuity, refractive status, and central corneal thickness) available from structured EHR data. One site was reserved as an "external site" test set (N = 1550); of the patients from the remaining sites, 10% each were randomly selected to be in development and test sets, with the remaining 27 999 reserved for model training. Main Outcome Measures: Evaluation metrics included area under the receiver operating characteristic curve (AUROC) on the test set and the external site. Results: Six thousand nineteen (16.5%) of 36 548 patients underwent glaucoma surgery. Overall, the AUROC ranged from 0.735 to 0.771 on the random test set and from 0.706 to 0.754 on the external test site, with the XGBoost and random forest model performing best, respectively. There was greatest performance decrease from the random test set to the external test site for the penalized regression models. Conclusions: Machine learning models developed using structured EHR data can reasonably predict whether glaucoma patients will need surgery, with reasonable generalizability to an external site. Additional research is needed to investigate the impact of protected class characteristics such as race or gender on model performance and fairness. Financial Disclosures: Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Using Natural Language Processing to Identify Different Lens Pathology in Electronic Health Records.

PURPOSE: Nearly all published ophthalmology-related Big Data studies rely exclusively on International Classification of Diseases (ICD) billing codes to identify patients with particular ocular conditions. However, inaccurate or nonspecific codes may be used. We assessed whether natural language processing (NLP), as an alternative approach, could more accurately identify lens pathology. DESIGN: Database study comparing the accuracy of NLP versus ICD billing codes to properly identify lens pathology. METHODS: We developed an NLP algorithm capable of searching free-text lens exam data in the electronic health record (EHR) to identify the type(s) of cataract present, cataract density, presence of intraocular lenses, and other lens pathology. We applied our algorithm to 17.5 million lens exam records in the Sight Outcomes Research Collaborative (SOURCE) repository. We selected 4314 unique lens-exam entries and asked 11 clinicians to assess whether all pathology present in the entries had been correctly identified in the NLP algorithm output. The algorithm's sensitivity at accurately identifying lens pathology was compared with that of the ICD codes. RESULTS: The NLP algorithm correctly identified all lens pathology present in 4104 of the 4314 lens-exam entries (95.1%). For less common lens pathology, algorithm findings were corroborated by reviewing clinicians for 100% of mentions of pseudoexfoliation material and 99.7% for phimosis, subluxation, and synechia. Sensitivity at identifying lens pathology was better for NLP (0.98 [0.96-0.99] than for billing codes (0.49 [0.46-0.53]). CONCLUSIONS: Our NLP algorithm identifies and classifies lens abnormalities routinely documented by eye-care professionals with high accuracy. Such algorithms will help researchers to properly identify and classify ocular pathology, broadening the scope of feasible research using real-world data.

Reliability of Intraocular Pressure Measurements in a Low-Contact Drive-Through Setting.

PRCIS: Drive-through intraocular pressure (IOP) measurement using iCare tonometry is a promising method of low-contact, high-throughput IOP monitoring. However, owing to its vulnerability to variable measurement technique and local air currents, the iCare may overestimate IOPs. PURPOSE: During the COVID-19 pandemic, a drive-through IOP measurement protocol using the iCare tonometer was established to facilitate low-contact monitoring of select glaucoma patients. As the iCare may be prone to error due to variable measurement technique and local air currents, we endeavored to assess the reliability of drive-through IOP measurements by comparing them with recent measurements taken in clinic settings. METHODS: Inclusion criteria were patients with drive-through IOP measurements performed from April 28 to October 11, 2020; exclusion criteria were pre-drive-through IOPs >21 mmHg. Drive-through IOP measurements were compared with the closest previous and/or subsequent in-clinic IOP measurements. Data were gathered using the Sight Outcomes Research Collaborative (SOURCE) data repository. RESULTS: The post-exclusion study group consisted of 314 patients receiving a total of 868 drive-through IOP measurements, all of whom had prior in-clinic measurements, and 56.8% of whom had subsequent in-clinic measurements. Drive-through IOPs were, on average, +2.4 mmHg (+14.5%; SD 4.9) higher than in-clinic IOPs. Further sub-analysis of the data showed a difference of +2.1 mmHg OD and +2.6 mmHg OS. Compared with the closest previous in-clinic visit, the difference was +2.4 mmHg OU (+2.1 mmHg OD, +2.7 mmHg OS); compared with the closest subsequent in-clinic visit, the difference was +2.3 mmHg OU (+2.1 mmHg OD, +2.5 mmHg OS). 68.6% of all drive-through IOPs were higher than corresponding in-clinic IOPs; 21.1% were lower. 25.9% of drive-through IOPs were higher by more than 5 mmHg, whereas 3.9% of drive-through IOPs were lower by more than 5 mmHg. DISCUSSION: As teleophthalmology becomes an ever more important tool in glaucoma patient care, drive-through or walk-through IOP monitoring methods are likely to play an increasing role. However, our data reveals potential inaccuracies in drive-through iCare IOP measurements which tended to overestimate IOP. It is advisable to confirm large changes in IOP with in-clinic measurement before making management decisions. CONCLUSION: With better optimization of accuracy and reliability of measurements, drive-through tonometry is a promising, high-throughput, low-contact method of measuring IOP.

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.

MAEPI and CIR: New Metrics for Robust Evaluation of the Prediction Performance of AI-Based IOL Formulas.

Purpose: To develop a class of new metrics for evaluating the performance of intraocular lens power calculation formulas robust to issues that can arise with AI-based methods. Methods: The dataset consists of surgical information and biometry measurements of 6893 eyes of 5016 cataract patients who received Alcon SN60WF lenses at University of Michigan's Kellogg Eye Center. We designed two types of new metrics: the MAEPI (Mean Absolute Error in Prediction of Intraocular Lens [IOL]) and the CIR (Correct IOL Rate) and compared the new metrics with traditional metrics including the mean absolute error (MAE), median absolute error, and standard deviation. We evaluated the new metrics with simulation analysis, machine learning (ML) methods, as well as existing IOL formulas (Barrett Universal II, Haigis, Hoffer Q, Holladay 1, PearlDGS, and SRK/T). Results: Results of traditional metrics did not accurately reflect the performance of overfitted ML formulas. By contrast, MAEPI and CIR discriminated between accurate and inaccurate formulas. The standard IOL formulas received low MAEPI and high CIR, which were consistent with the results of the traditional metrics. Conclusions: MAEPI and CIR provide a more accurate reflection of the real-life performance of AI-based IOL formula than traditional metrics. They should be computed in conjunction with conventional metrics when evaluating the performance of new and existing IOL formulas. Translational Relevance: The proposed new metrics would help cataract patients avoid the risks caused by inaccurate AI-based formulas, whose true performance cannot be determined by traditional metrics.

Ray tracing intraocular lens calculation performance improved by AI-powered postoperative lens position prediction.

AIMS: To assess whether incorporating a machine learning (ML) method for accurate prediction of postoperative anterior chamber depth (ACD) improves cataract surgery refraction prediction performance of a commonly used ray tracing power calculation suite (OKULIX). METHODS AND ANALYSIS: A dataset of 4357 eyes of 4357 patients with cataract was gathered at the Kellogg Eye Center, University of Michigan. A previously developed machine learning (ML)-based method was used to predict the postoperative ACD based on preoperative biometry measured with the Lenstar LS900 optical biometer. Refraction predictions were computed with standard OKULIX postoperative ACD predictions and ML-based predictions of postoperative ACD. The performance of the ray tracing approach with and without ML-based ACD prediction was evaluated using mean absolute error (MAE) and median absolute error (MedAE) in refraction prediction as metrics. RESULTS: Replacing the standard OKULIX postoperative ACD with the ML-predicted ACD resulted in statistically significant reductions in both MAE (1.7% after zeroing mean error) and MedAE (2.1% after zeroing mean error). ML-predicted ACD substantially improved performance in eyes with short and long axial lengths (p<0.01). CONCLUSIONS: Using an ML-powered postoperative ACD prediction method improves the prediction accuracy of the OKULIX ray tracing suite by a clinically small but statistically significant amount, with the greatest effect seen in long eyes.

Quality of life associated with no light perception vision.

OBJECTIVE: Time trade-off (TTO) utility analysis quantifies the quality of life associated with best-seeing-eye (BSE) vision. We compared the patient quality of life associated with unilateral and bilateral no light perception (NLP) with that of a control cohort without NLP. DESIGN: Cross-sectional interviews using a validated, reliable TTO vision utility analysis instrument. PARTICIPANTS: A total of 1598 consecutive ophthalmology patients from the authors' practices. METHODS: Patient records were reviewed in a case-control fashion The utilities of participants with unilateral or bilateral NLP vision were compared with those from patients without NLP vision. RESULTS: Among 99 NLP patients, 93 (94%) had unilateral NLP and 6 (6%) had bilateral NLP, for a total of 105 NLP eyes. Multiple regression analysis demonstrated the highest correlation between utility and BSE acuity (p = 0.001), with no correlation with age, ophthalmic disease, time of vision loss, race, or education. Mean unilateral NLP utility ranged from 0.55 in the counting fingers to light perception subcohort to 0.80 in the 20/20-20/25 subcohort. The 6-person bilateral NLP subcohort had a 0.54 utility. The 99-patient NLP cohort mean utility was 0.69, a 55% quality-of-life decrease versus a BSE vision-matched 0.80 in 1499 non-NLP patients (p < 0.001). CONCLUSIONS: TTO utility in unilateral NLP patients correlated with BSE vision at a lower utility than in patients with matched BSE vision without fellow-eye NLP. Decreased unilateral NLP patient quality of life should be considered in cost-utility analysis and clinical management. Bilateral NLP patient utility (0.54) was slightly less than that (0.55) in blind unilateral NLP patients with fellow-eye counting fingers to light perception vision, suggesting that more study is needed.

Augmenting Kalman Filter Machine Learning Models with Data from OCT to Predict Future Visual Field Loss: An Analysis Using Data from the African Descent and Glaucoma Evaluation Study and the Diagnostic Innovation in Glaucoma Study.

Purpose: To assess whether the predictive accuracy of machine learning algorithms using Kalman filtering for forecasting future values of global indices on perimetry can be enhanced by adding global retinal nerve fiber layer (RNFL) data and whether model performance is influenced by the racial composition of the training and testing sets. Design: Retrospective, longitudinal cohort study. Participants: Patients with open-angle glaucoma (OAG) or glaucoma suspects enrolled in the African Descent and Glaucoma Evaluation Study or Diagnostic Innovation in Glaucoma Study. Methods: We developed a Kalman filter (KF) with tonometry and perimetry data (KF-TP) and another KF with tonometry, perimetry, and global RNFL data (KF-TPO), comparing these models with one another and with 2 linear regression (LR) models for predicting mean deviation (MD) and pattern standard deviation values 36 months into the future for patients with OAG and glaucoma suspects. We also compared KF model performance when trained on individuals of European and African descent and tested on patients of the same versus the other race. Main Outcome Measures: Predictive accuracy (percentage of MD values forecasted within the 95% repeatability interval) differences among the models. Results: Among 362 eligible patients, the mean ± standard deviation age at baseline was 71.3 ± 10.4 years; 196 patients (54.1%) were women; 202 patients (55.8%) were of European descent, and 139 (38.4%) were of African descent. Among patients with OAG (n = 296), the predictive accuracy for 36 months in the future was higher for the KF models (73.5% for KF-TP, 71.2% for KF-TPO) than for the LR models (57.5%, 58.0%). Predictive accuracy did not differ significantly between KF-TP and KF-TPO (P = 0.20). If the races of the training and testing set patients were aligned (versus nonaligned), the mean absolute prediction error of future MD improved 0.39 dB for KF-TP and 0.48 dB for KF-TPO. Conclusions: Adding global RNFL data to existing KFs minimally improved their predictive accuracy. Although KFs attained better predictive accuracy when the races of the training and testing sets were aligned, these improvements were modest. These findings will help to guide implementation of KFs in clinical practice.

The Frequency of Visual Field Testing in a US Nationwide Cohort of Individuals with Open-Angle Glaucoma.

PURPOSE: Visual field testing that is not frequent enough results in delayed identification of open-angle glaucoma (OAG) progression. Guidelines recommend at least annual testing. It is not known how frequently patients with OAG across the United States receive visual field testing and how patient characteristics and circumstances influence this frequency. If US patients with OAG do not receive visual field tests frequently enough, interventions to increase this frequency or to develop other forms of testing visual function may reduce unidentified OAG vision loss. DESIGN: Retrospective cohort study. PARTICIPANTS: The TruvenHealth MarketScan Commercial Claims Database (IBM) contains demographic and claims data for > 160 million individuals across the United States from 2008 to 2017. We identified enrollees in the database with a recorded diagnosis of OAG (International Classification of Diseases, Ninth Revision, Clinical Modification and International Classification of Diseases, Tenth Revision, Clinical Modification codes 356.1x and H40.1x, respectively). We excluded those aged < 40 years at the time of their first OAG diagnosis, those without at least 1 confirmatory OAG diagnosis at a subsequent visit, and those with < 4 years of follow-up data after OAG diagnosis. METHODS: We calculated the number of visual field tests that each enrollee with OAG underwent per year and categorized the enrollees based on that number (0, > 0 to < 0.9, ≥ 0.9 to ≤ 1.1, > 1.1 to ≤ 2.1, and > 2.1). We used negative binomial regression to investigate the demographic or health variables that were associated with the frequency of visual field tests that enrollees with OAG received. MAIN OUTCOME MEASURES: Frequency of visual field testing among enrollees with OAG. RESULTS: Of the 380 029 enrollees included in the study, 33 267 (8.8%) did not receive a visual field test during the study period, 259 349 (68.2%) underwent > 0 to < 0.9 visual field tests per year, 42 129 (11.1%) underwent ≥ 0.9 to ≤ 1.1 visual field tests per year, 42 301 (11.1%) underwent > 1.1 to ≤ 2.1 visual field tests per year, and 2983 (0.8%) underwent ≥ 2.1 visual field tests per year. The median number of visual field tests per year was 0.63 (interquartile range, 0.33-0.88; mean, 0.65). CONCLUSIONS: More than 75% of enrollees with OAG received < 1 visual field test per year and, thus, did not receive guideline-adherent glaucoma monitoring.

Potentially Missed Opportunities in Prevention of Acute Angle-Closure Crisis.

Importance: If an anatomic narrow angle is not appropriately diagnosed and treated, it can result in acute angle-closure crisis (AACC) and lead to substantial vision loss. Objective: To identify patients who presented with AACC and assess for factors that may have been associated with risk of developing it. Design, Setting, and Participants: This population-based retrospective cohort study conducted from January 1, 2001, to December 31, 2015, included a 20% nationwide sample of 1179 Medicare beneficiaries. Patients aged 40 years or older with AACC were identified with billing codes. A 2-year lookback period from the date of initial presentation of AACC was used to identify patients who had at least 1 eye care visit, received a diagnosis of open-angle glaucoma (OAG) or suspected OAG, or received at least 1 medication associated with risk of AACC. Of the patients who had at least 1 eye care visit, those who underwent gonioscopy, received a diagnosis of an anatomic narrow angle before developing AACC, or both were identified. Main Outcomes and Measures: Proportions of patients who had at least 1 eye care visit, had OAG or suspected OAG, received at least 1 medication associated with risk of AACC, underwent gonioscopy, or received a diagnosis of an anatomic narrow angle before development of AACC. Results: A total of 1179 patients had a confirmed diagnosis of AACC. The mean (SD) age of patients with AACC was 66.7 (11.8) years (range, 40-96 years), 766 were women (65.0%), 57 were Asian (4.8%), 109 were Black (9.2%), 126 were Latino (10.7%), 791 were White (67.1%), and 96 were other race and ethnicity (8.1%). Of these patients, only 796 (67.5%) consulted an optometrist or ophthalmologist at least once during the 2-year lookback period. A total of 464 individuals (39.4%) had OAG or suspected OAG, and 414 (35.1%) had received at least 1 medication associated with increased risk of AACC before developing it. Of the 796 patients who consulted an optometrist or ophthalmologist in the lookback period, less than one-third underwent gonioscopy in the 2 years before developing AACC (n = 264 [33.2%]), and less than one-half of all patients undergoing gonioscopy received a diagnosis of an anatomic narrow angle (n = 113 [42.8%]). Most patients underwent gonioscopy in the 1 to 4 weeks preceding the AACC. Conclusions and Relevance: In this group of Medicare patients, there appear to have been multiple opportunities for interventions that may have averted AACC. Interventions aimed at addressing risk factors associated with AACC and improving performance of gonioscopy might be associated with reduced risk for ocular morbidity.

The Impact of COVID-19 and Pandemic Mitigation Measures on Persons With Sensory Impairment.

PURPOSE: To assess the impact of the COVID-19 pandemic and associated mitigation measures on persons with sensory impairments (SI), including visual impairments (VI) and hearing impairments (HI). DESIGN: Cross-sectional survey. METHODS: Adults with VI (best-corrected visual acuity <20/60 in the better-seeing eye), HI (International Classification of Diseases, Tenth Revision, codes), and age- and sex-matched controls (n = 375) were recruited from the University of Michigan. The 34-item Coronavirus Disability Survey was administered. Both χ2 tests and logistic regression were used to compare survey responses between groups. RESULTS: All groups reported high levels of disruption of daily life, with 80% reporting "a fair amount" or "a lot" of disruption (VI: 76%, HI: 83%, CT: 82%, P = .33). Participants with VI had greater difficulty with day-to-day activities and were more likely to cite the following reasons: caregiver was worried about COVID-19 (odds ratio [OR]VI = 7.2, 95% CI = 3.5-14.4, P < .001) and decreased availability of public transportation (ORVI = 5.0, 95% CI = 1.5-15.6, P = .006). Participants with VI, but not HI, showed a trend toward increased difficulty accessing medical care (ORVI = 2.0, 95% CI = 0.99-4.0, P = .052) and began relying more on others for day-to-day assistance (ORVI = 3.1, 95% CI = 1.6-5.7, P < .001). Overall, 30% reported difficulty obtaining trusted information about the pandemic. Those with VI reported more difficulty seeing or hearing trusted information (ORVI = 6.1, 95% CI = 1.6-22.1, P = .006). Employed participants with HI were more likely to report a reduction in wages (ORHI = 2.5, 95% CI = 1.2-5.3, P = .02). CONCLUSIONS: Individuals with VI have experienced increased disruption and challenges in daily activities related to the pandemic. People with SI may benefit from targeted policy approaches to the current pandemic and future stressors. Minimal differences in some survey measures may be due to the large impact of the pandemic on the population as a whole. The SARS-CoV-2 (COVID-19) pandemic and public health mitigation measures have had an exceedingly large impact around the globe. As of the time of writing, more than 114 million global cases (28 million US) had been diagnosed, and there had been more than 2.5 million fatalities attributed to COVID-19 (517,000 US).1,2.