Multitrait analysis of glaucoma identifies new risk loci and enables polygenic prediction of disease susceptibility and progression

The investigated collaborative eye health-care model led to a substantial improvement in appropriate referrals of glaucoma patients to ophthalmologists and could be suitable for optimizing patient care and utilization of resources. Improvement in follow-up of patients by optometrists is required to minimize inappropriately discontinued patient care.

No abstract

To compare the visual fields results obtained using the Swedish interactive thresholding algorithm-Standard (SS) and the Swedish interactive thresholding algorithm-Faster (SFR) in normal subjects, glaucoma suspects, and patients with glaucoma and to quantify potential time-saving benefits of the SFR algorithm. DESIGN: Prospective, cross-sectional study. METHODS: One randomly selected eye from 364 patients (77 normal subjects, 178 glaucoma suspects, and 109 patients with glaucoma) seen in a single institution underwent testing using both SS and SFR on the Humphrey Field Analyzer. Cumulative test time using each algorithm was compared after accounting for different rates of test reliability. Pointwise and cluster analysis was performed to determine whether there were systematic differences between algorithms. RESULTS: Using SFR had a greater rate of unreliable results (29.3%) compared with SS (7.7%, P < .0001). This was mainly because of high false positive rates and seeding point errors. However, modeled test times showed that using SFR could obtain a greater number of reliable results within a shorter period of time. SFR resulted in higher sensitivity values (on average 0.5 dB for patients with glaucoma) that was greater under conditions of field loss (<19 dB). Cluster analysis showed no systematic patterns of sensitivity differences between algorithms. CONCLUSIONS: After accounting for different rates of test reliability, SFR can result in significant time savings compared with SS. Clinicians should be cognizant of false positive rates and seeding point errors as common sources of error for SFR. Results between algorithms are not directly interchangeable, especially if there is a visual field deficit <19 dB.

Glaucoma, a leading cause of blindness, is a multifaceted disease with several patho-physiological features manifesting in single fundus images (e.g., optic nerve cupping) as well as fundus videos (e.g., vascular pulsatility index). Current convolutional neural networks (CNNs) developed to detect glaucoma are all based on spatial features embedded in an image. We developed a combined CNN and recurrent neural network (RNN) that not only extracts the spatial features in a fundus image but also the temporal features embedded in a fundus video (i.e., sequential images). A total of 1810 fundus images and 295 fundus videos were used to train a CNN and a combined CNN and Long Short-Term Memory RNN. The combined CNN/RNN model reached an average F-measure of 96.2% in separating glaucoma from healthy eyes. In contrast, the base CNN model reached an average F-measure of only 79.2%. This proof-of-concept study demonstrates that extracting spatial and temporal features from fundus videos using a combined CNN and RNN, can markedly enhance the accuracy of glaucoma detection.

No abstract

Pressure is a crucial component of the cellular environment, and can lead to pathology if it varies beyond its normal range. The increased intra-ocular pressures in acute glaucoma are associated with the loss of neurons by apoptosis. Little is known regarding the interaction between pressure and apoptosis at the level of the cell. The model developed in this study examines the effects of elevated ambient hydrostatic pressure directly upon cultured neuronal lines. Conditions were selected to be within physiological limits: 100 mmHg over and above atmospheric pressure for a period of 2 hr, as seen clinically in acute glaucoma. This system can be used to investigate pressure relatively independently of other variables. Neuronal cell line cultures (B35 and PC12) were subjected to pressure conditions in specially designed pressure chambers. Controls were treated identically, except for the application of pressure, and positive controls were treated with a known apoptotic stimulus. Apoptosis was detected by cell morphology changes and by 2 specific apoptotic markers: TUNEL (Terminal transferase dUTP Nick-End Labeling) and Annexin V. These fluorescent markers were detected and quantified by automated Laser Scanning Cytometry. All techniques showed that increased pressure was associated with a greater level of apoptosis compared to equivalent controls. Our results suggest that pressure alone may act as a stimulus for apoptosis in neuronal cell cultures. This raises the possibility of a more direct relationship at the cellular level between pressure and neuronal loss.

Objective: To examine the combined effects of common genetic variants associated with intraocular pressure (IOP) on primary open angle glaucoma (POAG) phenotype using a polygenic risk score (PRS) stratification. Design: Cross-sectional study. Participants: For the primary analysis, we examined the glaucoma phenotype of 2,154 POAG patients enrolled in the Australian and New Zealand Registry of Advanced Glaucoma (ANZRAG) including cases recruited from the UK. For replication, we examined an independent cohort of 624 early POAG patients. Methods: Using IOP genome-wide association study summary statistics, we developed a PRS derived solely from IOP associated variants and stratified POAG patients into three risk tiers. The lowest and highest quintiles of the score were set as the low and high risk groups respectively and the other quintiles as the intermediate risk group. Main Outcome Measures: Clinical glaucoma phenotype including maximum recorded IOP, age of diagnosis, number of family members affected by glaucoma, cup-to-disc ratio, visual field mean deviation, and treatment intensity. Results: There was a dose-response relationship between the IOP PRS and the maximum recorded IOP, with the high genetic risk group having a higher maximum IOP by 1.7 (SD 0.62) mmHg than the low genetic risk group (P = 0.006). Compared to the low genetic risk group, the high genetic risk group had a younger age of diagnosis by 3.7 (1.0) years (P < 0.001), more family members affected by 0.46 (0.11) members (P < 0.001), and higher rates of incisional surgery (odds ratio 1.5; 95% confidence interval 1.1-2.0; P = 0.007). There was no statistically significant difference in mean deviation. We further replicated the maximum IOP, number of family members affected by glaucoma and treatment intensity (number of medications) results in the early POAG cohort (P ≤ 0.01). Conclusions: The IOP polygenic risk score was positively correlated with maximum IOP, disease severity, need for surgery and number of family members. Genes acting via IOP mediated pathways, when considered in aggregate have clinically important and reproducible implications for glaucoma patients and their close family members.