Punctal Stenosis Associated with Topical Netarsudil Use.

PURPOSE: To report a series of patients who developed punctal stenosis secondary to the use of topical netarsudil 0.02% for treatment of glaucoma. DESIGN: Case series. PARTICIPANTS: Patients using topical netarsudil for management of glaucoma and noted to have punctal stenosis ipsilateral to the eye(s) being treated with netarsudil were included. METHODS: Each enrolled patient's chart was reviewed, and alternative causes of punctal stenosis were sought. Photographs were obtained to document punctal stenosis for some patients. MAIN OUTCOME MEASURES: Presence of punctal stenosis after topical netarsudil use and resolution of punctal stenosis after cessation of therapy. RESULTS: Sixteen patients had punctal stenosis; 13 developed unilateral punctal stenosis while using netarsudil unilaterally, and 3 patients developed bilateral punctal stenosis with bilateral use. Time from initiation of netarsudil to recognition of symptoms or documentation of punctal stenosis ranged from 2 to 35 months (median, 12; mean, 14.0 ± 8.7 months). Thirteen patients endorsed tearing, but 2 had no symptoms. Ectropion was seen in 1 eye. Corneal verticillata was noted in 14 patients (87.5%). In 8 cases, netarsudil was discontinued, and the punctal stenosis was reversed, with resolution of associated symptoms. CONCLUSIONS: Netarsudil use can lead to the development of reversible punctal stenosis. This inflammation-mediated stenosis may cause tearing and associated symptoms and may be of sufficient severity to necessitate discontinuation of treatment. In this case series, all patients who discontinued treatment had reversal of their punctal stenosis and associated symptoms.

Cyclodialysis cleft formation following Yamane secondary intraocular lens implantation.

PURPOSE: To report two cases of hypotony with maculopathy related to cyclodialysis cleft after Yamane intraocular lens (IOL) implantation, and the use of localized transscleral cyclophotocoagulation (TSCPC) to successfully treat this condition. OBSERVATION: 1. 37 year-old man with childhood BB-gun related injury in the left eye (OS) and traumatic iridodialysis and angle recession underwent pars plana vitrectomy and Yamane IOL placement for subluxed traumatic cataract OS. Postoperative hypotony [intraocular pressure (IOP) 5-6 mmHg] and maculopathy with best corrected vision acuity (BCVA) of 20/200 at 1 month postoperative prompted referral, and localized TSCPC was performed. Nine days later sudden elevation of IOP occurred, responsive to treatment, and the hypotony and maculopathy resolved. 2. 87 year-old man with prior OS retinal detachments treated with scleral buckling, pars plana vitrectomy x 2, and cataract extraction with sulcus IOL ranging from 8 to 37 years prior presented with temporal sulcus IOL haptic penetration through the iris and dense vitreous hemorrhage. He underwent pars plana vitrectomy, IOL explantation and Yamane IOL placement OS. Postoperative hypotony (IOP 1-4 mmHg) and maculopathy with evidence of cyclodialysis cleft on ultrasonography at 1 month postoperative prompted referral. The patient underwent 2 rounds of localized TSCPC; after his second treatment, IOP ranged from 9 to 14 mmHg over the next 8 months and maculopathy resolved. CONCLUSION AND IMPORTANCE: We highlight the risk of development of cyclodialysis cleft after Yamane IOL placement in highly traumatized eyes, and the benefit of localized TSCPC in such cases for cleft closure.

How should we measure intraocular pressure in the era of coronavirus disease 2019? Balancing infectious risk, cleaning requirements, and accuracy.

PURPOSE OF REVIEW: Accurate and precise measurement of intraocular pressure (IOP) is a vitally important component of the ophthalmic examination. There are multiple methods of tonometry, each of which has considerations in light of the ongoing severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) pandemic. This review discusses these considerations and compares various tonometer methods with the gold standard of Goldmann applanation tonometry (GAT). RECENT FINDINGS: The SARS-CoV-2 virus may spread via droplets, microaerosols, or direct contact in the ophthalmology clinic. Tonometry poses a high risk of contamination. The accuracy and reliability of various methods of tonometry with single-use disposable equipment has been compared with Goldmann applanation tonometry. SUMMARY: Goldmann applanation tonometry with disposable applanation tips, Tono-pen, and iCare employ single use tips to decrease the risk of cross-contamination of infectious agents. Review of the literature demonstrates good correlation between these devices and GAT, although the published level of agreement between devices varies.

Automated Detection of Glaucoma With Interpretable Machine Learning Using Clinical Data and Multimodal Retinal Images.

PURPOSE: To develop a multimodal model to automate glaucoma detection DESIGN: Development of a machine-learning glaucoma detection model METHODS: We selected a study cohort from the UK Biobank data set with 1193 eyes of 863 healthy subjects and 1283 eyes of 771 subjects with glaucoma. We trained a multimodal model that combines multiple deep neural nets, trained on macular optical coherence tomography volumes and color fundus photographs, with demographic and clinical data. We performed an interpretability analysis to identify features the model relied on to detect glaucoma. We determined the importance of different features in detecting glaucoma using interpretable machine learning methods. We also evaluated the model on subjects who did not have a diagnosis of glaucoma on the day of imaging but were later diagnosed (progress-to-glaucoma [PTG]). RESULTS: Results show that a multimodal model that combines imaging with demographic and clinical features is highly accurate (area under the curve 0.97). Interpretation of this model highlights biological features known to be related to the disease, such as age, intraocular pressure, and optic disc morphology. Our model also points to previously unknown or disputed features, such as pulmonary function and retinal outer layers. Accurate prediction in PTG highlights variables that change with progression to glaucoma-age and pulmonary function. CONCLUSIONS: The accuracy of our model suggests distinct sources of information in each imaging modality and in the different clinical and demographic variables. Interpretable machine learning methods elucidate subject-level prediction and help uncover the factors that lead to accurate predictions, pointing to potential disease mechanisms or variables related to the disease.

Accuracy of partial coherence interferometry in patients with large inter-eye axial length difference.

BACKGROUND: To determine accuracy of partial coherence interferometry (PCI) in patients with large inter-eye axial eye length (AEL) difference. METHODS: Patients undergoing cataract surgery at two academic medical centers with an inter-eye axial eye length (AEL) difference of > 0.30 mm were identified and were matched to control patients without inter-eye AEL difference > 0.30 mm on the basis of age, sex, and AEL. The expected post-operative refraction for the implanted IOL was calculated using SRK/T, Holladay II, and Hoffer Q formulae. The main outcome measures were the refractive prediction error and the equivalence of the refractive outcomes between the subjects and controls. RESULTS: Review of 2212 eyes from 1617 patients found 131 eyes of 93 patients which met inclusion criteria. These were matched to 131 control eyes of 115 patients. The mean AEL was 24.92 ± 1.50 mm. The mean absolute error (MAE) ranged from 0.47 D to 0.69 D, and was not statistically different between subjects and controls. The refractive prediction error was equivalent between the cases and controls, with no significant difference between the MAE for any formula, nor in the number of cases vs. controls with a refractive prediction error of at least 0.50 D or 1.00 D. CONCLUSIONS: Among eyes in our study population, good-quality PCI data was equally accurate in patients with or without an inter-eye AEL difference > 0.30 mm. Confirmatory AEL measurements using different AEL measuring modalities in patients with a large inter-eye AEL difference may not be necessary.

Incidence of and Risk Factors for Steroid Response After Cataract Surgery in Patients With and Without Glaucoma.

PRECIS: Steroid response after cataract surgery was more frequent in glaucoma patients than nonglaucoma patients. Longer axial length and more preoperative medications were risk factors for steroid response in glaucoma patients. PURPOSE: The aim was to evaluate incidence and risk factors for topical steroid response after uneventful cataract surgery in patients with and without glaucoma. SETTING: Academic glaucoma clinics. DESIGN: This was a retrospective review. PARTICIPANTS: Consecutive patients with and without glaucoma and no prior incisional glaucoma surgery undergoing cataract surgery between March 2007 and September 2016. All patients routinely received topical prednisolone acetate 1% postoperatively. METHODS: Pertinent clinical information was recorded. Steroid response was defined as intraocular pressure >50% above the baseline intraocular pressure measurement, occurring at or after the second postoperative week. RESULTS: We included 472 eyes of 472 nonglaucoma patients and 191 eyes of 191 glaucoma patients. Ten (2.1%) nonglaucoma eyes and 16 (8.4%) glaucoma eyes were diagnosed as steroid responders (relative risk=3.72; 95% confidence interval: 1.71-8.07; P<0.001). Logistic regression showed that for nonglaucoma, longer axial length (AL) and younger age were associated with a higher incidence of steroid response (P≤0.003), while for glaucoma patients, longer AL and more preoperative medications were associated with steroid response (P≤0.030). An AL ≥26 mm was associated with steroid response for both groups (P≤0.024). CONCLUSION: Although glaucoma patients were 3.72 times more likely to have steroid response after uneventful cataract surgery, the incidence of steroid response with prednisolone acetate 1% was relatively low after phacoemulsification in both nonglaucoma and glaucoma eyes. Steroid response was associated with longer AL in both groups and with more preoperative medications in glaucoma patients.

Macular microvascular parameters in the ganglion cell-inner plexiform layer derived by optical coherence tomography angiography: Vascular structure-central visual function analysis.

PURPOSE: To investigate the relationships between global and sectoral macular vascular microcirculation parameters in the ganglion cell-inner plexiform layer (GCIPL) assessed by optical coherence tomography angiography (OCTA), and global and sectoral visual field (VF) central mean sensitivity (CMS) assessed by standard automated perimetry. METHODS: Fifty-four eyes with open angle glaucoma were scanned using a swept-source OCTA (Plex Elite 9000, Zeiss, Dublin, CA) and macular vascular microcirculation was measured by calculating the overall flux and vessel area density (VAD) over the entire 6mm x 6mm area, excluding large retinal vessels. Central 10-degree VF CMS was calculated based on 24-2 VF. Pearson correlation was used to investigate the correlation between global and sectoral OCTA parameters and global and sectoral VF CMS. RESULTS: Both global GCIPL flux and VAD were significantly correlated with VF CMS (p<0.001). For the sectoral analysis, sectoral VAD was significantly correlated with sectoral VF CMS in all comparisons except for the inferonasal VF CMS with supero-temporal (ST) GCIPL VAD (p = 0.097). Although highest correlation was observed for both ST VF CMS with inferior GCIPL VAD and infero-temporal VF CMS with superior GCIPL VAD (r = 0.683, p<0.001), there was no significant difference in correlation when compared to the global VAD and other sectors' correlation coefficients (p≥ 0.091), except for the ST GCIPL VAD (p = 0.001). CONCLUSIONS: Global and sectoral macular vascular microcirculation in the GCIPL, as determined by OCTA, was significantly correlated with global and sectoral VF CMS in glaucomatous patients. OCTA can aid in the understanding of the structure-function relationships of the macular region.

Peripapillary and Macular Microcirculation in Glaucoma Patients of African and European Descent Using Optical Coherence Tomography Angiography.

PRéCIS:: We found no significant differences in peripapillary and macula microcirculation blood flow metrics in eyes with open-angle glaucoma of African descent (AD) and European descent (ED) as detected by optical coherence tomography angiography (OCTA). PURPOSE: The purpose of this study was to investigate the peripapillary retinal nerve fiber layer (RNFL) and macular vascular microcirculation in subjects of AD and ED with open-angle glaucoma using OCTA. PATIENTS AND METHODS: One eye from each subject was scanned using AngioPlex OCTA system covering both a 6×6 mm scanning area centered at the optic nerve head and at the foveola. Peripapillary RNFL and macular microcirculation were measured by calculating the overall flux and vessel area density excluding the large retinal vessels. Two-sample, independent t tests were used to compare the OCTA metrics between AD and ED eyes. Linear regression models were used to investigate the correlation between OCTA metrics and structural and functional parameters. RESULTS: Twenty-eight eyes of AD and 56 eyes of ED were included in the study. There was no significant difference in age, sex, hypertension, antihypertensive medications, diabetes, systolic and diastolic blood pressure, mean ocular perfusion pressure, RNFL thickness and visual field (VF) mean deviation and VF pattern standard deviation (P≥0.054) between AD and ED eyes included. Both groups had similar OCTA blood flow metrics (P≥0.161). OCTA blood flow metrics were significantly correlated with VF mean deviation (r≥0.466), VF pattern standard deviation (r≤-0.366) and RNFL thickness (r≥0.333). CONCLUSIONS: No significant differences were found in peripapillary and macular microcirculation detected by OCTA between AD and ED glaucomatous eyes. Peripapillary and macular microcirculation were significantly correlated with disease severity in AD and ED glaucomatous eyes.

Pseudophakic adult with progressive optic disc tilt and axial length elongation.

PURPOSE: To describe a case of progressive optic disc tilt in a pseudophakic Asian adult associated with an increase in axial length. OBSERVATIONS: A 28-year-old man presented with high myopia in the right eye with an axial length of 30.4 mm and refraction of -13.75 diopters (D) spherical equivalent (SE). The refraction was plano sphere after LASIK and LASIK enhancement. The left eye was blinded by retinal detachment after phacoemulsification. Six years later, the right eye axial length was 31.65 mm when he underwent phacoemulsification with intraocular lens placement for nuclear sclerosis, after which the refraction was -0.50 D SE. The right eye refraction changed progressively over 13 years after cataract surgery to -4.00 D SE, at which time the axial length was 33.69 mm. A notable increase in optic disc tilt and peripapillary atrophy was evident on comparison of fundus photographs taken 11 years apart. CONCLUSIONS: Although progressive optic disc tilt in conjunction with increasing axial myopia in pediatric populations has been documented, it has not been reported in adults but may occur. Scleral stretching due to axial length elongation associated with loss of accommodation after cataract surgery and worsening peripapillary atrophy are possible causes of this finding.

Using Deep Learning to Automate Goldmann Applanation Tonometry Readings.

PURPOSE: To develop an objective and automated method for measuring intraocular pressure using deep learning and fixed-force Goldmann applanation tonometry (GAT) techniques. DESIGN: Prospective cross-sectional study. PARTICIPANTS: Patients from an academic glaucoma practice. METHODS: Intraocular pressure was estimated by analyzing videos recorded using a standard slit-lamp microscope and fixed-force GAT. Video frames were labeled to identify the outline of the reference tonometer and the applanation mires. A deep learning model was trained to localize and segment the tonometer and mires. Intraocular pressure values were calculated from the deep learning-predicted tonometer and mire diameters using the Imbert-Fick formula. A separate test set was collected prospectively in which standard and automated GAT measurements were collected in random order by 2 independent masked observers to assess the deep learning model as well as interobserver variability. MAIN OUTCOME MEASURES: Intraocular pressure measurements between standard and automated methods were compared. RESULTS: Two hundred sixty-three eyes of 135 patients were included in the training and validation videos. For the test set, 50 eyes from 25 participants were included. Each eye was measured by 2 observers, resulting in 100 videos. Within the test set, the mean difference between automated and standard GAT results was -0.9 mmHg (95% limits of agreement [LoA], -5.4 to 3.6 mmHg). Mean difference between the 2 observers using standard GAT was 0.09 mmHg (LoA,-3.8 to 4.0 mmHg). Mean difference between the 2 observers using automated GAT videos was -0.3 mmHg (LoA, -4.1 to 3.5 mmHg). The coefficients of repeatability for automated and standard GAT were 3.8 and 3.9 mmHg, respectively. The bias for even-numbered measurements was reduced when using automated GAT. CONCLUSIONS: Preliminary measurements using deep learning to automate GAT demonstrate results comparable with those of standard GAT. Automated GAT has the potential to improve on our current GAT measurement standards significantly by reducing bias and improving repeatability. In addition, ocular pulse amplitudes could be observed using this technique.

Visual Field Outcomes in the Tube Versus Trabeculectomy Study.

PURPOSE: To describe visual field (VF) outcomes in the Tube Versus Trabeculectomy (TVT) Study. DESIGN: Cohort analysis of patients in a multicenter randomized clinical trial. PARTICIPANTS: A total of 122 eyes of 122 patients, with 61 eyes in both the tube shunt and trabeculectomy groups. METHODS: The TVT Study is a multicenter randomized clinical trial comparing the safety and efficacy of tube shunt surgery (350-mm2 Baerveldt implant) and trabeculectomy with mitomycin C (MMC) (0.4 mg/ml for 4 minutes) in patients with previous cataract or glaucoma surgery. Enrolled patients underwent perimetry at baseline and annual follow-up visits. The VFs were included if the false-positive rate was ≤20% and false-negative rate was ≤35%. The VFs were excluded if visual acuity <20/400 or loss of ≥2 Snellen lines from baseline was attributed to an etiology other than glaucoma. Longitudinal linear mixed-effects models with best linear unbiased predictions (BLUPs) were applied to estimate rates of change in mean deviation (MD) for each treatment group. MAIN OUTCOME MEASURE: Rate of MD change during follow-up period. RESULTS: A total of 436 reliable VFs were analyzed, with an average of 3.6 VFs per eye. Baseline MD was -13.07 ± 8.4 decibels (dB) in the tube shunt group and -13.18 ± 8.2 dB in the trabeculectomy group (P = 0.99). The rate of change in MD was -0.60 dB/year in the tube group and -0.38 dB/year in the trabeculectomy group (P = 0.34). The 95% confidence intervals for the rates of MD change were -0.77 to -0.44 dB/year in the tube group and -0.56 to -0.20 dB/year in the trabeculectomy group. No significant difference in MD slope was seen when patients were categorized by percentage of visits with intraocular pressure (IOP) <18 mmHg or by average IOP. Univariable and multivariable risk factor analyses identified history of diabetes, elevated IOP, and worse MD as baseline factors associated with more rapid VF loss. CONCLUSIONS: Slow rates of VF loss were observed after randomized surgical treatment in the TVT Study, but no significant difference in the rate of VF loss was seen after tube shunt implantation and trabeculectomy with MMC. Patients with diabetes, higher IOP, and more severe VF loss at baseline were at higher risk for VF progression.

Development and validation of a machine learning, smartphone-based tonometer.

BACKGROUND/AIMS: To compare intraocular pressure (IOP) measurements using a prototype smartphone tonometer with other tonometers used in clinical practice. METHODS: Patients from an academic glaucoma practice were recruited. The smartphone tonometer uses fixed force applanation and in conjunction with a machine-learning computer algorithm is able to calculate the IOP. IOP was also measured using Goldmann applanation tonometry (GAT) in all subjects. A subset of patients were also measured using ICare, pneumotonometry (upright and supine positions) and Tono-Pen (upright and supine positions) and the results were compared. RESULTS: 92 eyes of 81 subjects were successfully measured. The mean difference (in mm Hg) for IOP measurements of the smartphone tonometer versus other devices was +0.24 mm Hg for GAT, -1.39 mm Hg for ICare, -3.71 mm Hg for pneumotonometry and -1.30 mm Hg for Tono-Pen. The 95% limits of agreement for the smartphone tonometer versus other devices was -4.35 to 4.83 mm Hg for GAT, -6.48 to 3.70 mm Hg for ICare, -7.66 to -0.15 mm Hg for pneumotonometry and -5.72 to 3.12 mm Hg for Tono-Pen. Overall, the smartphone tonometer results correlated best with GAT (R2=0.67, p<0.001). Of the 92 videos, 90 (97.8%) were within ±5 mm Hg of GAT and 58 (63.0%) were within ±2 mm Hg of GAT. CONCLUSIONS: Preliminary IOP measurements using a prototype smartphone-based tonometer was grossly equivalent to the reference standard.

Macular Vascular Microcirculation in Eyes With Open-angle Glaucoma Using Different Visual Field Severity Classification Systems.

PRECIS: We found significant differences in macular vascular microcirculation between normal and glaucomatous eyes using optical coherence tomography angiography (OCTA). Macular vascular microcirculation changes also showed significant correlations with visual field (VF) severity classification systems. PURPOSE: To correlate VF severity defined by different classification systems and macular vascular microcirculation in eyes with glaucoma using OCTA. PATIENTS AND METHODS: Twenty normal and 58 open-angle glaucoma (OAG) eyes were scanned using a swept-source OCTA (Plex Elite 9000) and macular vascular microcirculation was measured by calculating the overall blood flux index (BFI) and vessel area density (VAD) over the entire 6×6 mm area excluding the big retinal vessels. Glaucomatous eyes were staged into severity groups based on 4 VF severity classifications: Hodapp-Parrish-Anderson scale, Glaucoma Severity Staging system, ICD-10 glaucoma staging definitions, and VF mean deviation. Central 10-degree VF mean sensitivity (CMS) was calculated based on 24-2 VF. One-way analysis of variance was used to analyze the differences and correlation between macular vascular microcirculation and other clinical parameters. RESULTS: Glaucomatous eyes had significantly lower ganglion cell and inner plexiform layer BFI and VAD (P<0.0001) compared with normal eyes. In OAG patients, BFI and VAD were significantly higher in mild OAG compared with severe OAG with all VF disease severity classification systems (P<0.001). Glaucoma Severity Staging had the highest correlation with changes in macular vascular microcirculation metrics (r=0.734 for BFI; r=0.647 for VAD) and VF CMS had highest correlation with macular vascular microcirculation metrics (r=0.887 for BFI; r=0.903 for VAD). CONCLUSION: Macular vascular microcirculation metrics detected by OCTA correlate with disease severity in glaucomatous eyes. VF CMS, calculated from only 12 tested central 10-degree points, correlated best with macular OCTA.

Forecasting future Humphrey Visual Fields using deep learning.

PURPOSE: To determine if deep learning networks could be trained to forecast future 24-2 Humphrey Visual Fields (HVFs). METHODS: All data points from consecutive 24-2 HVFs from 1998 to 2018 were extracted from a university database. Ten-fold cross validation with a held out test set was used to develop the three main phases of model development: model architecture selection, dataset combination selection, and time-interval model training with transfer learning, to train a deep learning artificial neural network capable of generating a point-wise visual field prediction. The point-wise mean absolute error (PMAE) and difference in Mean Deviation (MD) between predicted and actual future HVF were calculated. RESULTS: More than 1.7 million perimetry points were extracted to the hundredth decibel from 32,443 24-2 HVFs. The best performing model with 20 million trainable parameters, CascadeNet-5, was selected. The overall point-wise PMAE for the test set was 2.47 dB (95% CI: 2.45 dB to 2.48 dB), and deep learning showed a statistically significant improvement over linear models. The 100 fully trained models successfully predicted future HVFs in glaucomatous eyes up to 5.5 years in the future with a correlation of 0.92 between the MD of predicted and actual future HVF and an average difference of 0.41 dB. CONCLUSIONS: Using unfiltered real-world datasets, deep learning networks show the ability to not only learn spatio-temporal HVF changes but also to generate predictions for future HVFs up to 5.5 years, given only a single HVF.