Progressive Changes in the Anterior Segment and Their Impact on the Anterior Chamber Angle in Primary Angle Closure Disease.

PURPOSE: To investigate longitudinal changes in the anterior segment (AS) using serial optical coherence tomography (OCT) images and determine the impact of these changes on the anterior chamber angle (ACA) in eyes with primary angle closure disease (PACD) treated with laser peripheral iridotomy (LPI). DESIGN: Retrospective clinical cohort study. METHODS: This study included 103 patients with PACD who underwent LPI and were followed up by a mean 6.7 ± 1.7 AS-OCT examinations for a mean 6.5 ± 2.9 years. Temporal changes in AS-OCT parameters, including anterior chamber depth (ACD), angle opening distance (AOD750), angle recess area (ARA750), iris thickness (IT750), lens vault (LV), and pupil diameter (PD), were analyzed by multivariate linear mixed effects models (LMEMs). RESULTS: Multivariate LMEMs showed that decrease in AOD750 was not significant (-1.59 µm/y, P = .222); however, ARA750 decreased over time (-2.3 × 103 µm2/y, P = .033) and SSA showed marginal significance (-0.20°/y, P = .098), and LV increased significantly (11.6 µm/y, P < .001) after LPI. Mean LV change was negatively associated with AOD750, ARA750, and SSA, whereas PD was negatively associated with ARA750 (P < .001 each). PD decreased with aging (-13.7 µm/y, P = .036), accompanied by thinning of IT750 (-1.7 µm/y, P = .063). CONCLUSIONS: LV tends to increase with aging, which contributes to the shallowing of the anterior chamber and narrowing of ACA in PACD eyes treated with LPI. In the meantime, pupillary constriction and subsequent peripheral iris thinning associated with aging could possibly offset the effect of ACA narrowing.

Development of a novel knockout model of retinitis pigmentosa using Pde6b-knockout Long-Evans rats.

Although rats with melanin-pigmentated retinal pigment epithelial (RPE) cells are physiologically more appropriate models for human eye research than their albino counterparts, reliable models from the former strain are not available to study retinal degeneration. Here, we describe the development of a novel Pde6b-knockout Long-Evans (LE Pde6b KO) rat model that recapitulates key features of human retinitis pigmentosa (RP). After the generation of the Pde6b-knockout Sprague-Dawley rats with the CRISPR-Cpf1 system, the LE rat was back-crossed over 5 generations to develop the pigmented LE Pde6b KO strain. Interestingly, LE Pde6b KO displayed well-developed bone-spicule pigmentation; a hallmark of fundus in patients with RP which cannot be observed in non-pigmented albino rats. Moreover, the rat model showed progressive thinning of the retina, which was evident by intravital imaging with optical coherence tomography. Histologically, significant atrophy was observed in the outer nuclear layer. Functionally, LE Pde6b KO presented a marked decrease of amplitude level during electroretinogram testing, demonstrating significant loss of visual function. Therefore, these findings suggest that the LE Pde6b KO model robustly recapitulates the hallmark phenotype of RP. We believe that the LE Pde6b KO model may be used effectively for preclinical translational research to further study retinal degeneration.

Development of Cumulative Order-Preserving Image Transformation Based Variational Autoencoder for Anterior Segment Optical Coherence Tomography Images.

Purpose: To develop a variational autoencoder (VAE) suitable for analysis of the latent structure of anterior segment optical coherence tomography (AS-OCT) images and to investigate possibilities of latent structure analysis of the AS-OCT images. Methods: We retrospectively collected clinical data and AS-OCT images from 2111 eyes of 1261 participants from the ongoing Asan Glaucoma Progression Study. A specifically modified VAE was used to extract six symmetrical and one asymmetrical latent variable. A total of 1692 eyes of 1007 patients were used for training the model. Conventional measurements and latent variables were compared between 74 primary angle closure (PAC) and 51 primary angle closure glaucoma (PACG) eyes from validation set (419 eyes of 254 patients) that were not used for training. Results: Among the symmetrical latent variables, the first three and the last demonstrated easily recognized features, anterior chamber area in η1, curvature of the cornea in η2, the pupil size in η3 and corneal thickness in η6, whereas η4 and η5 were more complex aggregating complex interactions of multiple structures. Compared with PAC eyes, there was no difference in any of the conventional measurements in PACG eyes. However, values of η4 were significantly different between the two groups, being smaller in the PACG group (P = 0.015). Conclusions: VAE is a useful framework for analysis of the latent structure of AS-OCT. Latent structure analysis could be useful in capturing features not readily evident with conventional measures. Translational Relevance: This study suggested that a deep learning-based latent space model can be applied for the analysis of AS-OCT images to find latent characteristics of the anterior segment of the eye.

Relationship Between Tamsulosin Use and Surgical Complications of Cataract Surgery in Elderly Patients: Population-Based Cohort Study.

Purpose: Although several previous studies have investigated the relationship between tamsulosin use and surgical complications of cataract surgery, no population-based cohort study has been conducted for the Asian population. We aimed to investigate the relationship between tamsulosin use and surgical complications of cataract surgery in the Korean elderly population. Methods: This nationwide population-based retrospective cohort study included elderly patients (≥60 years) who had undergone cataract surgery in the period from 2003 to 2015. Baseline characteristics were age, sex, income, residence, and systemic, and ocular comorbidities (glaucoma, myopia, eye trauma, diabetes mellitus with ophthalmic manifestations, severe cataract, age-related macular degeneration). The exposure of interest was tamsulosin use within 1 year before cataract surgery. Logistic regression model was used to evaluate the relationship of tamsulosin use with surgical complications of cataract surgery. Results: The rate of surgical complications of cataract surgery was 0.88% (375/42,539) in the non-tamsulosin group and 0.83% (71/8,510) in the tamsulosin group. The groups showed no significant difference in the risk of surgical complications of cataract surgery in the unadjusted model [odds ratio (OR) = 0.946; 95% confidence interval (CI):0.733-1.220; P = 0.669]. Additionally, tamsulosin use was not significantly associated with surgical complications of cataract surgery in the fully adjusted model accounting for age, income, residence, and systemic and ocular comorbidities (OR = 0.997; 95% CI: 0.749-1.325; P = 0.981). Conclusions: The rate or risk of surgical complications of cataract surgery does not change with tamsulosin use. We suggest that better surgical techniques and surgeons' cognizance of the patient's tamsulosin use could improve surgical outcomes, without increasing surgical complications.

Development of a ß-Variational Autoencoder for Disentangled Latent Space Representation of Anterior Segment Optical Coherence Tomography Images.

PURPOSE: To investigate the feasibility of extracting a low-dimensional latent structure of anterior segment optical coherence tomography (AS-OCT) images by use of a ß-variational autoencoder (ß-VAE). METHODS: We retrospectively collected 2111 AS-OCT images from 2111 eyes of 1261 participants from the ongoing Asan Glaucoma Progression Study. After hyperparameter optimization, the images were analyzed with ß-VAE. RESULTS: The mean participant age was 64.4 years, with mean values of visual field index and mean deviation of 86.4% and -5.33 dB, respectively. After experiments, a latent space size of 6 and ß value of 53 were selected for latent space analysis with ß-VAE. Latent variables were successfully disentangled, showing readily interpretable distinct characteristics, such as the overall depth and area of the anterior chamber (η1), pupil diameter (η2), iris profile (η3 and η4), and corneal curvature (η5). CONCLUSIONS: ß-VAE can successfully be applied for disentangled latent space representation of AS-OCT images, revealing the high possibility of applying unsupervised learning in the medical image analysis. TRANSLATIONAL RELEVANCE: This study demonstrates that a deep learning-based latent space model can be applied for the analysis of AS-OCT images.