Faculty Bibliography

PURPOSE/OBJECTIVE:The aims of this study were to determine rates of early postkeratoplasty endophthalmitis and identify sociodemographic and medical risk factors in the Medicare population. METHODS:Using a retrospective cohort design, patients aged 65 years and older undergoing penetrating keratoplasty (PK), endothelial keratoplasty (EK), and anterior lamellar keratoplasty (ALK) from 2016 to 2019 among 100% Medicare Fee-or-Service database were included. Rates of early endophthalmitis within 42 days of keratoplasty were determined using the International Classification of Diseases, 10th Revision-Clinical Modification diagnostic codes. Patient and physician characteristics were compared using x2 tests, and a multivariable logistic regression model was used to evaluate factors associated with endophthalmitis. RESULTS:The overall early endophthalmitis rate after keratoplasty was 0.39% (n = 216/54,822) with a median time to diagnosis of 14 (interquartile range: 5-25) days. Rates by keratoplasty types were 1.31% for cataract surgery combined with PK, 1.13% for PK, and 0.22% for EK. On multivariable analysis, the odds of endophthalmitis were higher for PK [odds ratio (OR): 5.46, 95% confidence interval (CI), 3.98-7.49] and ALK (OR: 5.45, 95% CI, 2.59-11.49) relative to EK. Patients with a Charlson Comorbidity Index (CCI) ≥3 had higher odds of endophthalmitis (OR: 1.82; 95% CI, 1.28-2.58) relative to patients with a CCI of 0. Practices located in the Midwest (OR: 0.59, 95% CI, 0.36-0.96), West (OR 0.57; 95% CI, 0.35-0.93), and Northeast (OR: 0.59 95% CI, 0.35-0.99) had lower odds of reporting endophthalmitis when compared to the South. CONCLUSIONS:Patients undergoing PK and ALK and those with a CCI ≥3 had higher odds of endophthalmitis relative to EK and patients without comorbidities, respectively. Practices in the West, Midwest, and Northeast had lower odds of endophthalmitis relative to the South.

PURPOSE/OBJECTIVE:To determine how often ChatGPT is able to provide accurate and comprehensive information regarding clinical vitreoretinal scenarios. To assess the types of sources ChatGPT primarily utilizes and to determine if they are hallucinated. METHODS:A retrospective cross-sectional study. We designed 40 open-ended clinical scenarios across 4 main topics in vitreoretinal disease. Responses were graded on correctness and comprehensiveness by two blinded retina specialists. The primary outcome was the number of clinical scenarios that ChatGPT answered correctly and comprehensively. Secondary outcomes included: theoretical harm to patients, the distribution of the type of references utilized by the chatbot, and the frequency of hallucinated references. RESULTS:In June 2023, ChatGPT answered 83% (33/40) of clinical scenarios correctly but provided a comprehensive answer in only 52.5% (21/40) of cases. Subgroup analysis demonstrated an average correct score of 86.7% in nAMD, 100% in DR, 76.7% in retinal vascular disease and 70% in the surgical domain. There were 6 incorrect responses with 1 (16.7%) case of no harm, 3 (50%) cases of possible harm and 2 (33.3%) cases of definitive harm. CONCLUSION/CONCLUSIONS:ChatGPT correctly answered more than 80% of complex open-ended vitreoretinal clinical scenarios, with a reduced capability to provide a comprehensive response.

PURPOSE/UNASSIGNED:We sought to introduce the materials, design, and biocompatibility of a flexible and suturable artificial corneal device. METHODS/UNASSIGNED:Single-piece, fully synthetic, optic-skirt design devices were made from compact perfluoroalkoxy alkane. The skirt and the optic wall surfaces were lined with a porous tissue ingrowth material using expanded polytetrafluoroethylene. Full-thickness macroapertures around the skirt perimeter were placed to facilitate nutrition of the recipient cornea. Material properties including the skirt's modulus of elasticity and bending stiffness, optic light transmission, wetting behavior, topical drug penetrance, and degradation profile were evaluated. RESULTS/UNASSIGNED:The final prototype suitable for human use has a transparent optic with a diameter of 4.60 mm anteriorly, 4.28 mm posteriorly, and a skirt outer diameter of 6.8 mm. The biomechanical and optical properties of the device closely align with the native human cornea with an average normalized device skirt-bending stiffness of 4.7 kPa·mm4 and light transmission in the visible spectrum ranging between 92% and 96%. No optical damage was seen in the 36 devices tested in fouling experiments. No significant difference was observed in topical drug penetrance into the anterior chamber of the device implanted eye compared with the naïve rabbit eye. CONCLUSIONS/UNASSIGNED:The flexibility and biocompatibility of our artificial cornea device may offer enhanced tissue integration and decreased inflammation, leading to improved retention compared with rigid keratoprosthesis designs. TRANSLATIONAL RELEVANCE/UNASSIGNED:We have developed a fully synthetic, flexible, suturable, optic-skirt design prototype artificial cornea that is ready to be tested in early human feasibility studies.

Some anterior chambers do not readily shallow because of insufficient posterior pressure and/or very deep anterior chamber anatomy, which can make unscrolling descemet membrane endothelial keratoplasty (DMEK) tissue more challenging with an unmodified tap technique. We present a hands-free method for augmenting posterior pressure by temporarily tucking cellulose sponges under the blades of the eyelid speculum. The sponges transfer some of the eyelid speculum's weight onto the bulbar surface posterior to the iris, thereby indenting the sclera and causing the iris diaphragm to bulge further forward. This hands-free technique can transform a potentially challenging DMEK case into a more straightforward one by facilitating both a shallow anterior chamber and a bimanual unscrolling technique. However, it only works in bicameral eyes with a vitreous body (e.g., an eye with penetrating keratoplasty, vitreous syneresis, and axial myopia) and will not work in unicameral eyes after vitrectomy (e.g., an eye with an Anterior Chamber Intraocular Lens (ACIOL)).

BACKGROUND AND OBJECTIVE/UNASSIGNED:The objective was to evaluate factors associated with clinical presentation of uveal melanoma (UM) during the initial two years of the coronavirus disease 2019 pandemic. PATIENTS AND METHODS/UNASSIGNED:This was a multi-site, retrospective cohort study of patients treated for uveal melanoma during the first (early) and second (late) year of the pandemic compared with the year prior (control). RESULTS/UNASSIGNED:≤ 0.0338) compared to the control cohort. CONCLUSIONS/UNASSIGNED:.

BACKGROUND:Prediction of drug-induced long QT syndrome (diLQTS) is of critical importance given its association with torsades de pointes. There is no reliable method for the outpatient prediction of diLQTS. OBJECTIVES/OBJECTIVE:This study sought to evaluate the use of a convolutional neural network (CNN) applied to electrocardiograms (ECGs) to predict diLQTS in an outpatient population. METHODS:We identified all adult outpatients newly prescribed a QT-prolonging medication between January 1, 2003, and March 31, 2022, who had a 12-lead sinus ECG in the preceding 6 months. Using risk factor data and the ECG signal as inputs, the CNN QTNet was implemented in TensorFlow to predict diLQTS. RESULTS:Models were evaluated in a held-out test dataset of 44,386 patients (57% female) with a median age of 62 years. Compared with 3 other models relying on risk factors or ECG signal or baseline QTc alone, QTNet achieved the best (P < 0.001) performance with a mean area under the curve of 0.802 (95% CI: 0.786-0.818). In a survival analysis, QTNet also had the highest inverse probability of censorship-weighted area under the receiver-operating characteristic curve at day 2 (0.875; 95% CI: 0.848-0.904) and up to 6 months. In a subgroup analysis, QTNet performed best among males and patients ≤50 years or with baseline QTc <450 ms. In an external validation cohort of solely suburban outpatient practices, QTNet similarly maintained the highest predictive performance. CONCLUSIONS:An ECG-based CNN can accurately predict diLQTS in the outpatient setting while maintaining its predictive performance over time. In the outpatient setting, our model could identify higher-risk individuals who would benefit from closer monitoring.

Early identification of Alzheimer's disease (AD) and AD-related dementias (ADRD) has high clinical significance, both because of the potential to slow decline through initiating FDA-approved therapies and managing modifiable risk factors, and to help persons living with dementia and their families to plan before cognitive loss makes doing so challenging. However, substantial racial and ethnic disparities in early diagnosis currently lead to additional inequities in care, urging accurate and inclusive risk assessment programs. In this study, we trained an artificial intelligence foundation model to represent the electronic health records (EHR) data with a vast cohort of 1.2 million patients within a large health system. Building upon this foundation EHR model, we developed a predictive Transformer model, named TRADE, capable of identifying risks for AD/ADRD and mild cognitive impairment (MCI), by analyzing the past sequential visit records. Amongst individuals 65 and older, our model was able to generate risk predictions for various future timeframes. On the held-out validation set, our model achieved an area under the receiver operating characteristic (AUROC) of 0.772 (95% CI: 0.770, 0.773) for identifying the AD/ADRD/MCI risks in 1 year, and AUROC of 0.735 (95% CI: 0.734, 0.736) in 5 years. The positive predictive values (PPV) in 5 years among individuals with top 1% and 5% highest estimated risks were 39.2% and 27.8%, respectively. These results demonstrate significant improvements upon the current EHR-based AD/ADRD/MCI risk assessment models, paving the way for better prognosis and management of AD/ADRD/MCI at scale.