Evaluations of aqueous humor protein markers in different types of glaucoma.

To compare the concentrations of protein markers in aqueous humor (AH) of patients with primary open-angle glaucoma (POAG), chronic angle-closure glaucoma (CACG), acute primary angle closure (APAC), and cataract without glaucoma as the control group. AH samples were collected at the beginning of surgery from 82 eyes of 82 patients who were divided into POAG (n = 23), CACG (n = 21), APAC (n = 19), and cataract groups (n = 19). The expression levels of interferon-gamma (IFN-γ), interleukin-2 (IL-2), interleukin-6 (IL-6), interleukin-8 (IL-8), interleukin-17A (IL-17A), lymphotoxin-alpha (LT-α), monocyte chemotactic protein-1 (MCP-1), matrix metalloproteinase-2 (MMP-2), brain derived neurotrophic factor (BDNF), basic fibroblast growth factor (bFGF), platelet-derived growth factor-AA (PDGF-AA), vascular endothelial growth factor (VEGF), tissue inhibitor of metalloproteinases-1 (TIMP-1), and tumor necrosis factor-alpha (TNF-α) in AH were detected using a microsphere-based immunoassay. The AH levels of TNF-α, MMP-2, MCP-1, IFN-γ, and TIMP-1 in the APAC and CACG groups were significantly higher than those in control eyes. Additionally, the AH levels of interleukin-6 (IL-6) and VEGF in the APAC group were significantly higher than those in the control group (CG). The interleukin-8 (IL-8) levels in patients with POAG were significantly higher than those in control eyes, whereas the LT-α levels were significantly lower than those in control eyes. IL-6 levels were significantly correlated with the coefficient of variation (CV), whereas IL-6 levels were significantly negatively correlated with the frequency of hexagonal cells (HEX) and corneal endothelial cell density (CD). The levels of TNF-α, MMP-2, MCP-1, IFN-γ, TIMP-1, IL-6, IL-8, VEGF, and LT-α were different among the three types of glaucoma. These different types of glaucoma may be caused by various pathogeneses, which opens avenues for further investigation into the pathogenesis of glaucoma and discoveries new targets and pathways for the treatment of glaucoma.

A Novel Context Aware Joint Segmentation and Classification Framework for Glaucoma Detection.

Glaucoma is a chronic ocular disease characterized by damage to the optic nerve resulting in progressive and irreversible visual loss. Early detection and timely clinical interventions are critical in improving glaucoma-related outcomes. As a typical and complicated ocular disease, glaucoma detection presents a unique challenge due to its insidious onset and high intra- and interpatient variabilities. Recent studies have demonstrated that robust glaucoma detection systems can be realized with deep learning approaches. The optic disc (OD) is the most commonly studied retinal structure for screening and diagnosing glaucoma. This paper proposes a novel context aware deep learning framework called GD-YNet, for OD segmentation and glaucoma detection. It leverages the potential of aggregated transformations and the simplicity of the YNet architecture in context aware OD segmentation and binary classification for glaucoma detection. Trained with the RIGA and RIMOne-V2 datasets, this model achieves glaucoma detection accuracies of 99.72%, 98.02%, 99.50%, and 99.41% with the ACRIMA, Drishti-gs, REFUGE, and RIMOne-V1 datasets. Further, the proposed model can be extended to a multiclass segmentation and classification model for glaucoma staging and severity assessment.

Real-World Clinical Impact of Netarsudil 0.02% at an Urban Safety-Net Hospital.

Purpose: To analyze the efficacy, safety, and accessibility of netarsudil 0.02% in patients with glaucoma (suspect, open or closed) at a safety-net academic medical center, Boston Medical Center (BMC). Methods: Retrospective chart review of patients prescribed netarsudil 0.02% for uncontrolled glaucoma at BMC between December 2017 and September 2019. Outcome measures included change in intraocular pressure (IOP) from baseline and evaluation of adverse events (AEs). Results: One hundred thirty patients (60% severe stage) were analyzed. The IOP reduction from baseline was about 3 mmHg. Fifty-four patients (42%) experienced an AE (eg, conjunctival hyperemia). Thirty-eight patients (29%) started netarsudil 0.02% in lieu of laser or surgery. Ninety-nine patients (71%) required prior authorization for insurance coverage of netarsudil 0.02%. Ten patients (7%) were unable to obtain netarsudil 0.02% due to issues with insurance coverage. Conclusion: Netarsudil 0.02% yielded significant IOP reduction in our cohort, however, to a smaller degree compared with prior studies that bore equivocal IOP reduction regardless of baseline IOP. Conjunctival hyperemia was the most common AE. In a limited number of patients, netarsudil 0.02% was not covered by insurance.

Glaucoma.

Glaucoma is the leading cause of irreversible blindness worldwide. The global prevalence of glaucoma in people aged 40 to 80 years is estimated to be 3.5%. With the growing number and proportion of older persons in the population, it is projected that 111.8 million people will have glaucoma in 2040. Currently available treatments cannot reverse glaucomatous damage to the visual system; however, early diagnosis and treatment can prevent progression of the disease. In most cases, glaucoma is a chronic condition that requires lifelong management. This article reviews the pathophysiology, classification, clinical manifestations, diagnosis, and management of glaucoma.

Childhood and Early Onset Glaucoma Classification and Genetic Profile in a Large Australasian Disease Registry.

PURPOSE: To report the relative frequencies of childhood and early onset glaucoma subtypes and their genetic findings in a large single cohort. DESIGN: Retrospective clinical and molecular study. PARTICIPANTS: All individuals with childhood glaucoma (diagnosed 0 to <18 years) and early onset glaucoma (diagnosed 18 to <40 years) referred to a national disease registry. METHODS: We retrospectively reviewed the referrals of all individuals with glaucoma diagnosed at <40 years of age recruited to the Australian and New Zealand Registry of Advanced Glaucoma (ANZRAG). Subtypes of glaucoma were determined using the Childhood Glaucoma Research Network (CGRN) classification system. DNA extracted from blood or saliva samples underwent sequencing of genes associated with glaucoma. MAIN OUTCOME MEASURES: The phenotype and genotype distribution of glaucoma diagnosed at <40 years of age. RESULTS: A total of 290 individuals (533 eyes) with childhood glaucoma and 370 individuals (686 eyes) with early onset glaucoma were referred to the ANZRAG. Primary glaucoma was the most prevalent condition in both cohorts. In the childhood cohort, 57.6% of individuals (167/290, 303 eyes) had primary congenital glaucoma (PCG), and 19.3% (56/290, 109 eyes) had juvenile open-angle glaucoma. Juvenile open-angle glaucoma constituted 73.2% of the early onset glaucoma cohort (271/370, 513 eyes). Genetic testing in probands resulted in a diagnostic yield of 24.7% (125/506) and a reclassification of glaucoma subtype in 10.4% of probands (13/125). The highest molecular diagnostic rate was achieved in probands with glaucoma associated with nonacquired ocular anomalies (56.5%). Biallelic variants in CYP1B1 (n = 29, 23.2%) and heterozygous variants in MYOC (n = 24, 19.2%) and FOXC1 (n = 21, 16.8%) were most commonly reported among probands with a molecular diagnosis. Biallelic CYP1B1 variants were reported in twice as many female individuals as male individuals with PCG (66.7% vs. 33.3%, P = 0.02). CONCLUSIONS: We report on the largest cohort of individuals with childhood and early onset glaucoma from Australasia using the CGRN classification. Primary glaucoma was most prevalent. Genetic diagnoses ascertained in 24.7% of probands supported clinical diagnoses and genetic counseling. International collaborative efforts are required to identify further genes because the majority of individuals still lack a clear molecular diagnosis.

Deep Learning Ensemble Method for Classifying Glaucoma Stages Using Fundus Photographs and Convolutional Neural Networks.

Purpose: This study developed and evaluated a deep learning ensemble method to automatically grade the stages of glaucoma depending on its severity.Materials and Methods: After cross-validation of three glaucoma specialists, the final dataset comprised of 3,460 fundus photographs taken from 2,204 patients were divided into three classes: unaffected controls, early-stage glaucoma, and late-stage glaucoma. The mean deviation value of standard automated perimetry was used to classify the glaucoma cases. We modeled 56 convolutional neural networks (CNN) with different characteristics and developed an ensemble system to derive the best performance by combining several modeling results.Results: The proposed method with an accuracy of 88.1% and an average area under the receiver operating characteristic of 0.975 demonstrates significantly better performance to classify glaucoma stages compared to the best single CNN model that has an accuracy of 85.2% and an average area under the receiver operating characteristic of 0.950. The false negative is the least adjacent misprediction, and it is less in the proposed method than in the best single CNN model.Conclusions: The method of averaging multiple CNN models can better classify glaucoma stages by using fundus photographs than a single CNN model. The ensemble method would be useful as a clinical decision support system in glaucoma screening for primary care because it provides high and stable performance with a relatively small amount of data.

The trabecular meshwork: Structure, function and clinical implications. A review of the literature.

Glaucoma is a blinding optic neuropathy, the main risk factor for which is increased intraocular pressure (IOP). The trabecular meshwork, located within the iridocorneal angle, is the main pathway for drainage of aqueous humor (AH) out of the eye, and its dysfunction is responsible for the IOP elevation. The trabecular meshwork is a complex, fenestrated, three-dimensional structure composed of trabecular meshwork cells (TMC) interdigitated into a multilayered organization within the extracellular matrix (ECM). The purpose of this literature review is to provide an overview of current understanding of the trabecular meshwork and its pathophysiology in glaucoma. Thus, we will present the main anatomical and cellular bases for the regulation of aqueous humor outflow resistance, the pathophysiological mechanisms involved in trabecular dysfunction in the various types of glaucoma, as well as current and future therapeutic strategies targeting the trabecular meshwork.

Deep learning assisted detection of glaucomatous optic neuropathy and potential designs for a generalizable model.

PURPOSE: To evaluate ways to improve the generalizability of a deep learning algorithm for identifying glaucomatous optic neuropathy (GON) using a limited number of fundus photographs, as well as the key features being used for classification. METHODS: A total of 944 fundus images from Taipei Veterans General Hospital (TVGH) were retrospectively collected. Clinical and demographic characteristics, including structural and functional measurements of the images with GON, were recorded. Transfer learning based on VGGNet was used to construct a convolutional neural network (CNN) to identify GON. To avoid missing cases with advanced GON, an ensemble model was adopted in which a support vector machine classifier would make final classification based on cup-to-disc ratio if the CNN classifier had low-confidence score. The CNN classifier was first established using TVGH dataset, and then fine-tuned by combining the training images of TVGH and Drishti-GS datasets. Class activation map (CAM) was used to identify key features used for CNN classification. Performance of each classifier was determined through area under receiver operating characteristic curve (AUC) and compared with the ensemble model by diagnostic accuracy. RESULTS: In 187 TVGH test images, the accuracy, sensitivity, and specificity of the CNN classifier were 95.0%, 95.7%, and 94.2%, respectively, and the AUC was 0.992 compared to the 92.8% accuracy rate of the ensemble model. For the Drishti-GS test images, the accuracy of the CNN, the fine-tuned CNN and ensemble model was 33.3%, 80.3%, and 80.3%, respectively. The CNN classifier did not misclassify images with moderate to severe diseases. Class-discriminative regions revealed by CAM co-localized with known characteristics of GON. CONCLUSIONS: The ensemble model or a fine-tuned CNN classifier may be potential designs to build a generalizable deep learning model for glaucoma detection when large image databases are not available.

Glaucoma classification based on intra-class and extra-class discriminative correlation and consensus ensemble classifier.

Automatic classification of glaucoma from fundus images is a vital diagnostic tool for Computer-Aided Diagnosis System (CAD). In this work, a novel fused feature extraction technique and ensemble classifier fusion is proposed for diagnosis of glaucoma. The proposed method comprises of three stages. Initially, the fundus images are subjected to preprocessing followed by feature extraction and feature fusion by Intra-Class and Extra-Class Discriminative Correlation Analysis (IEDCA). The feature fusion approach eliminates between-class correlation while retaining sufficient Feature Dimension (FD) for Correlation Analysis (CA). The fused features are then fed to the classifiers namely Support Vector Machine (SVM), Random Forest (RF) and K-Nearest Neighbor (KNN) for classification individually. Finally, Classifier fusion is also designed which combines the decision of the ensemble of classifiers based on Consensus-based Combining Method (CCM). CCM based Classifier fusion adjusts the weights iteratively after comparing the outputs of all the classifiers. The proposed fusion classifier provides a better improvement in accuracy and convergence when compared to the individual algorithms. A classification accuracy of 99.2% is accomplished by the two-level hybrid fusion approach. The method is evaluated on the public datasets High Resolution Fundus (HRF) and DRIVE datasets with cross dataset validation.

Mood and behavior seasonality in glaucoma; assessing correlations between seasonality and structure and function of the retinal ganglion cells.

AIM: In glaucoma, depression and disturbed sleep has been associated with degeneration of the intrinsically photosensitive retinal ganglion cells, that mediate non-image forming effects of light such as regulation of circadian rhythm, alertness and mood. In this study we assessed associations between seasonal mood and behavior variation and retinal ganglion cell damage in outpatients with glaucoma. METHODS: The seasonal pattern assessment questionnaire was administered to outpatients with glaucoma. Data on visual field defects identified by autoperimetry and retinal nerve fiber layer thickness visualized by ocular coherence tomography were collected from patient charts. The correlations between seasonality and retinal damage were tested and the adjusted effects of retinal function on seasonality were evaluated in a linear regression model. RESULTS: In total, 113 persons completed the questionnaire. Of these, 4% fulfilled the criteria for seasonal affective disorder (SAD) and 8% for subsyndromal seasonal affective disorder (sSAD). Mean global seasonal score was 4.3. There were no significant correlations between seasonality and either visual field or retinal nerve fiber layer thickness. In the adjusted analysis there were trends toward differential effects of visual field on seasonality in subgroups with different sex and type of glaucoma. CONCLUSION: There were no strong associations between seasonality and visual field or retinal nerve fiber layer thickness. Sex, age and glaucoma subtype may modify light effects on complex regulatory systems.

Diagnosing Glaucoma With Spectral-Domain Optical Coherence Tomography Using Deep Learning Classifier.

PRéCIS:: A spectral-domain optical coherence tomography (SD-OCT) based deep learning system detected glaucomatous structural change with high sensitivity and specificity. It outperformed the clinical diagnostic parameters in discriminating glaucomatous eyes from healthy eyes. PURPOSE: The purpose of this study was to assess the performance of a deep learning classifier for the detection of glaucomatous change based on SD-OCT. METHODS: Three hundred fifty image sets of ganglion cell-inner plexiform layer (GCIPL) and retinal nerve fiber layer (RNFL) SD-OCT for 86 glaucomatous eyes and 307 SD-OCT image sets of 196 healthy participants were recruited and split into training (197 eyes) and test (85 eyes) datasets based on a patient-wise split. The bottleneck features extracted from the GCIPL thickness map, GCIPL deviation map, RNFL thickness map, and RNFL deviation map were used as predictors for the deep learning classifier. The area under the receiver operating characteristic curve (AUC) was calculated and compared with those of conventional glaucoma diagnostic parameters including SD-OCT thickness profile and standard automated perimetry (SAP) to evaluate the accuracy of discrimination for each algorithm. RESULTS: In the test dataset, this deep learning system achieved an AUC of 0.990 [95% confidence interval (CI), 0.975-1.000] with a sensitivity of 94.7% and a specificity of 100.0%, which was significantly larger than the AUCs with all of the optical coherence tomography and SAP parameters: 0.949 (95% CI, 0.921-0.976) with average GCIPL thickness (P=0.006), 0.938 (95% CI, 0.905-0.971) with average RNFL thickness (P=0.003), and 0.889 (0.844-0.934) with mean deviation of SAP (P<0.001; DeLong test). CONCLUSION: An SD-OCT-based deep learning system can detect glaucomatous structural change with high sensitivity and specificity.

Facilitating Patient-Ophthalmologist Dialog: A Call for a Patient-focused Classification of Glaucoma Diagnosis.

We present a recommended patient-oriented glaucoma classification to facilitate patient-ophthalmologist dialog. By improving patients' understanding of their precise diagnosis, we hope to optimize management outcomes. We invite our colleagues to evolve this classification with us.

Artificial Intelligence Classification of Central Visual Field Patterns in Glaucoma.

PURPOSE: To quantify the central visual field (VF) loss patterns in glaucoma using artificial intelligence. DESIGN: Retrospective study. PARTICIPANTS: VFs of 8712 patients with 13 951 Humphrey 10-2 test results from 13 951 eyes for cross-sectional analyses, and 824 patients with at least 5 reliable 10-2 test results at 6-month intervals or more from 1191 eyes for longitudinal analyses. METHODS: Total deviation values were used to determine the central VF patterns using the most recent 10-2 test results. A 24-2 VF within a 3-month window of the 10-2 tests was used to stage eyes into mild, moderate, or severe functional loss using the Hodapp-Anderson-Parrish scale at baseline. Archetypal analysis was applied to determine the central VF patterns. Cross-validation was performed to determine the optimal number of patterns. Stepwise regression was applied to select the optimal feature combination of global indices, average baseline decomposition coefficients from central VFs archetypes, and other factors to predict central VF mean deviation (MD) slope based on the Bayesian information criterion (BIC). MAIN OUTCOME MEASURES: The central VF patterns stratified by severity stage based on 24-2 test results and a model to predict the central VF MD change over time using baseline test results. RESULTS: From cross-sectional analysis, 17 distinct central VF patterns were determined for the 13 951 eyes across the spectrum of disease severity. These central VF patterns could be divided into isolated superior loss, isolated inferior loss, diffuse loss, and other loss patterns. Notably, 4 of the 5 patterns of diffuse VF loss preserved the less vulnerable inferotemporal zone, whereas they lost most of the remaining more vulnerable zone described by the Hood model. Inclusion of coefficients from central VF archetypical patterns strongly improved the prediction of central VF MD slope (BIC decrease, 35; BIC decrease of >6 indicating strong prediction improvement) than using only the global indices of 2 baseline VF results. Eyes with baseline VF results with more superonasal and inferonasal loss were more likely to show worsening MD over time. CONCLUSIONS: We quantified central VF patterns in glaucoma, which were used to improve the prediction of central VF worsening compared with using only global indices.

Utility of Optical Coherence Tomography (OCT) in Centers For Medicare and Medicaid Services (CMS)-defined Severe Glaucoma Patients.

PRECIS: Patients with Centers For Medicare and Medicaid Services (CMS)-defined severe glaucoma often have clinically useful remaining retinal nerve fiber layer (RNFL), suggesting that structurally based rather than functionally based criteria would be more appropriate to use as guidelines for the utilization of optical coherence tomography (OCT) imaging. PURPOSE: RNFL OCT in glaucomatous eyes with advanced structural damage can reach a floor after which there is no further detectable thinning of RNFL. Insurers are considering limiting coverage for OCT in "severe stage glaucoma" defined by CMS. However, CMS definitions of severe glaucoma are based primarily on visual field criteria. Many of these patients may have preserved RNFL in other sectors. This study aims to assess the clinical utility of RNFL measurements by OCT in eyes with CMS-defined severe glaucoma. PATIENTS AND METHODS: Medical records of patients with CMS-defined severe glaucoma were consecutively reviewed. Data collected included average/sectoral RNFL thickness, and visual field mean deviation. Previous estimates of RNFL floor and test-retest variability for Cirrus OCT were used to establish 3 threshold values of RNFL. Data analysis included descriptive statistics, χ analysis, and analysis of variance. RESULTS: A total of 129 eyes qualified (age: 71±12 y; mean deviation: -13.5±4.3 dB; average RNFL: 60.9±7.9 µm), A majority (66%) of eyes met severe glaucoma criteria with defects in both hemifields; 34% met only paracentral defect criteria. The proportion of eyes that had significant remaining average, superior, or inferior RNFL, estimated by thresholds 1 to 3, was 21% to 54%, 29% to 51%, and 16% to 37%, respectively. At least 1 vertical quadrant had significant remaining RNFL in 35% to 66% of eyes, depending on the threshold used. CONCLUSIONS: Our data demonstrate a substantial portion of eyes with CMS-defined severe glaucoma have measurable RNFL above the floor in at least 1 vertical quadrant that may be longitudinally monitored for progression. The presence of CMS-defined severe glaucoma does not exclude the potential utility of OCT to monitor progression.

Risk Factors for Choroidal Detachment After Ahmed Valve Implantation in Glaucoma Patients.

PURPOSE: To assess the incidence and risk factors related to choroidal detachment after glaucoma drainage device (Ahmed valve) implantation. DESIGN: Retrospective case-control study. METHODS: A total of 188 eyes of 188 glaucoma patients were enrolled who underwent Ahmed valve implantation surgery. Patients were divided into 2 groups according to the presence or absence of choroidal detachment. The data were analyzed for factors associated with choroidal detachment. Separately, we divided eyes with choroidal detachment into 2 subgroups according to severity and conducted a subanalysis. In addition, we also analyzed the factors associated with chamber collapse. RESULTS: The incidence of choroidal detachment was 35.1% according to wide-field fundus photography and 16.9% according to 45-degree fundus photography. The current study showed that age, central corneal thickness, axial length, etiology of glaucoma, history of cataract or glaucoma, hypertension, diabetes, and severity of the visual field (MD) were different between the choroidal detachment and nonchoroidal detachment groups. A multivariate analysis showed significant differences in age (P = .035), etiology of glaucoma (pseudoexfoliation; PEX) (P = .028), lens status (pseudophakia) (P = .011), and hypertension (P = .011). The greater the intraocular pressure difference before and after surgery, the greater the size of the choroidal detachment. Chamber collapse risk was associated with only short axial length. CONCLUSION: The detection of choroidal detachment after Ahmed valve implantation can be increased according to the introduction of wide fundus photography. The risk of choroidal detachment is associated with the etiology of glaucoma (PEX), older age, pseudophakia (lens status), and hypertension.

A feature agnostic approach for glaucoma detection in OCT volumes.

Optical coherence tomography (OCT) based measurements of retinal layer thickness, such as the retinal nerve fibre layer (RNFL) and the ganglion cell with inner plexiform layer (GCIPL) are commonly employed for the diagnosis and monitoring of glaucoma. Previously, machine learning techniques have relied on segmentation-based imaging features such as the peripapillary RNFL thickness and the cup-to-disc ratio. Here, we propose a deep learning technique that classifies eyes as healthy or glaucomatous directly from raw, unsegmented OCT volumes of the optic nerve head (ONH) using a 3D Convolutional Neural Network (CNN). We compared the accuracy of this technique with various feature-based machine learning algorithms and demonstrated the superiority of the proposed deep learning based method. Logistic regression was found to be the best performing classical machine learning technique with an AUC of 0.89. In direct comparison, the deep learning approach achieved a substantially higher AUC of 0.94 with the additional advantage of providing insight into which regions of an OCT volume are important for glaucoma detection. Computing Class Activation Maps (CAM), we found that the CNN identified neuroretinal rim and optic disc cupping as well as the lamina cribrosa (LC) and its surrounding areas as the regions significantly associated with the glaucoma classification. These regions anatomically correspond to the well established and commonly used clinical markers for glaucoma diagnosis such as increased cup volume, cup diameter, and neuroretinal rim thinning at the superior and inferior segments.

Association between radiation, glaucoma subtype, and retinal vessel diameter in atomic bomb survivors.

We examined the relationship between glaucoma subtype and retinal vascular caliber as markers of ocular circulation. Subjects were Japanese atomic bomb survivors in Hiroshima and Nagasaki. After a screening examination, potential cases were subjected to further definitive examination. The diameters of central retinal artery and vein equivalents (CRAE and CRVE) on digitized retinal photographs were measured using an established method. Generalized linear regression analyses were used to examine the associations among vessel diameters, radiation exposure, and prevalence of glaucoma subtypes among the study subjects. We identified 196 cases of glaucoma (12%) based on optic disc appearance, perimetry results, and other ocular findings. The main subtypes were primary angle-closure glaucoma, primary open-angle glaucoma and normal-tension glaucoma (NTG). NTG was the dominant subtype (78%). NTG was negatively associated with CRAE and CRVE, and positively associated with radiation dose. CRVE was negatively associated with radiation dose and the association was unclear for CRAE. The smaller retinal vessel caliber in NTG patients than in subjects without glaucoma may indicate an association between ocular blood flow and the pathogenesis of NTG. However, significant relationships among vessel calibers, NTG and radiation exposure were not clear.

Frequencies of 4 Distinct Patterns of Glaucomatous Disc Appearance and Their Clinical Associations in Japanese Population-based Studies.

PURPOSE: To study the frequencies and factors associated with 4 disc patterns in primary open-angle glaucoma (POAG) identified in population-based studies: focal glaucomatous (FG type), generalized enlargement of cup (GE type), myopic glaucomatous (MG type), and senile sclerotic glaucomatous (SS type) patterns. SUBJECTS: In total, 270 disc photographs of acceptable quality were extracted from the records of 270 definitive POAG cases diagnosed according to the International Society of Geographical and Epidemiological Ophthalmology Criteria in 2 Japanese population-based glaucoma surveys. One randomly chosen eye from the bilateral POAG cases was included. RESULTS: Using a method of κ coefficient of reproducibility of classification of 0.80 according to a preliminary study, 143 discs were classified as FG, GE, MG, or SS types with respective frequencies of 57% (95% confidence interval [CI], 48-66), 33% (95% CI, 25-42), 7% (95% CI, 3-13), and 3% (95% CI, 0-7), and 127 discs as the miscellaneous type. Multinomial logistic regression analysis showed that the MG type was associated (P=0.052, 0.025, 0.019, and 0.018) with younger age, lower body mass index (BMI), and greater disc area and ovality, and the GE type was associated (P<0.001, 0.036, and 0.056) with greater disc area, corneal radius, and hyperopic refraction than the FG type. CONCLUSIONS: The FG type occurs most frequently in Japanese POAG followed by the GE type. The MG and SS types occurred much less often than previously reported in Japanese. Associations with age, BMI, disc area and ovality, refraction, and corneal radius differed among the FG, GE, and MG types.