Association Between Perforating Scleral Vessel and Myopic Maculopathy: A Cross-Sectional Study of a Chinese Cohort

Purpose: To investigate the association between perforating scleral vessel (PSV) and different types of myopic maculopathy (MM) in a highly myopic population.Methods: In total, 188 highly myopic eyes (117 participants) were enrolled. Each participant underwent detailed history taking and ocular examinations. Based on fundus photographs and optical coherence tomography, patients were subdivided into the non-MM group and MM group. Based on a new classification system (ATN), MM cases were classified as myopic atrophy maculopathy (MAM), myopic tractional maculopathy (MTM), and myopic neovascular maculopathy (MNM). The number of PSV and the macular choroidal thickness (mChT) were measured.Results: Compared with non-MM group, MM group was characterized by relatively larger age (48.40 vs. 32.34; p < 0.001), longer axial length (AL, 29.72 vs. 27.75, p < 0.001), thinner mChT (52.90 vs. 122.52; p < 0.001), and lower PSV counts (6.73 vs. 9.47, p ≤ 0.001). The non-MM group had higher PSV counts in total area (0–9 mm, 9.47 vs. 6.73, p < 0.001) and perifovea area (3–9 mm, 7.25 vs. 4.71, p < 0.001) compared to the MM group. Univariate and multivariate analyses showed that PSV count had no association with MAM (p = 0.2419) and MTM (p = 0.5678). Total PSV count [odds ratio (OR) 0.78, 95% CI 0.64–0.95, p = 0.0149] and perifovea PSV count (OR 0.80, 95% CI 0.65–0.98, p = 0.0299) were both protective factors for MNM. The stratified analysis revealed that in groups with AL <28 mm, or mChT <50 μm, or mChT ≥100 μm, or eyes with cilioretinal artery, PSV count had no significant association with MNM.Conclusion: Higher PSV counts in perifovea area (3–9 mm centered fovea) and total area (0–9 mm centered fovea) were protective factors for MNM, whereas PSV count had no association with MAM and MTM. These findings may provide novel insights into the mechanisms of pathologic myopia.

Novel BMP4 Truncations Resulted in Opposite Ocular Anomalies: Pathologic Myopia Rather Than Microphthalmia

BMP4 variants have been reported to be associated with syndromic microphthalmia (MCOPS6, OMIM 607932). This study aims to describe BMP4 truncation mutations contributing to a novel phenotype in eight patients from four Chinese families. In this study, BMP4 variants were collected from a large dataset from in-house exome sequencing. Candidate variants were filtered by multiple in silico tools as well as comparison with data from multiple databases. Potential pathogenic variants were further confirmed by Sanger sequencing and cosegregation analysis. Four novel truncation variants in BMP4 were detected in four out of 7,314 unrelated probands with different eye conditions. These four mutations in the four families solely cosegregated in all eight patients with a specific form of pathologic myopia, characterized by significantly extended axial length, posterior staphyloma, macula patchy, chorioretinal atrophy, myopic optic neuropathy or glaucoma, vitreous opacity, and unique peripheral snow-grain retinopathy. The extreme rarity of the truncations in BMP4 (classified as intolerant in the gnomAD database, pLI = 0.96), the exclusive presence of these variants in the four families with pathologic myopia, variants fully co-segregated with the same specific phenotypes in eight patients from the four families, and the association of the pathogenicity of truncations with syndromic microphthalmia in previous studies, all support a novel association of BMP4 truncations with a specific form of pathologic myopia. The data presented in this study demonstrated that heterozygous BMP4 truncations contributed to a novel phenotype: pathologic myopia rather than microphthalmia. Mutations in the same gene resulting in both high myopia and microphthalmia have been observed for a few other genes like FZD5 and PAX6, suggesting bidirectional roles of these genes in early ocular development. Further studies are expected to elucidate the molecular mechanism of the bidirectional regulation.

A Rare Case of Chronic Primary Angle-Closure Glaucoma in a Young Woman with Coexisting Pathologic Myopia

Introduction: Primary Angle-Closure Glaucoma (PACG) is usually present in adults older than forty and is more common in hyperopic eyes. Angle-closure is usually related to structural or developmental ocular abnormalities in young individuals. Case presentation: We presented a rare case of PACG in a 32 years old woman with pathologic myopia of -23.0 RLE. The chief complaint was blurring of the right eye three months before the visit. Right eye Intraocular Pressure (IOP) was 30mmHg-38mmHg despite treatment with three intra-ocular pressure-lowering agents. Axial length was 32.36 mm and 31.19 mm RLE. Anterior chamber depth was 2.36 mm and 2.60 mm RLE. Lens thickness was 5.07 mm and 5.40 mm RLE. Signs of GON and pathologic myopia were found in both eyes. GON was present asymmetrically (0.9 and 0.6 RLE), with the myopic crescent as well as baring and peripapillary atrophy. The optic disc was slightly tilted with the myopic crescent. There was also a marked sign of retinal pigment epithelium thinning and attenuation along with myopic chorioretinal atrophy. Conclusions: PACG in a young myopic individual is challenging to diagnose because myopia and glaucoma share similar optic nerve head pathology. Comprehensive examinations including gonioscopy, biometry, and OCT may confirm the diagnosis. In the presented case, angle-closure was caused by thick lenses and a shallow anterior chamber, along with excessively long axial length. Primary angle-closure at a young individual with myopic eyes is highly uncommon. Nonetheless, clinicians should always consider glaucoma even in the presence of high axial length and myopic fundus.

Morphology and Microcirculation Changes of Optic Nerve Head Between Simple High Myopia and Pathologic Myopia

This study explored morphology and microcirculation changes of optic nerve head (ONH) in simple high myopia(SHM) and pathologic myopia(PM), in order to evaluate and identify ONH changes in the development of PM. We divided 193 right eyes of 193 patients into SHM and PM according to the retinopathy. We found that ONH is one of the earliest pathological changes in myopia, and its morphology changes were also the most obvious. PM is closely linked to the reduction of choroidal perfusion and structural changes of ONH. Microcirculation showed a significant priority changes in myopia. Further research should address whether these fndings are associated with future disease development in highly myopic eyes.

Development of deep learning-based detecting systems for pathologic myopia using retinal fundus images

Globally, cases of myopia have reached epidemic levels. High myopia and pathological myopia (PM) are the leading cause of visual impairment and blindness in China, demanding a large volume of myopia screening tasks to control the rapid growing myopic prevalence. It is desirable to develop the automatically intelligent system to facilitate these time- and labor- consuming tasks. In this study, we designed a series of deep learning systems to detect PM and myopic macular lesions according to a recent international photographic classification system (META-PM) classification based on color fundus images. Notably, our systems recorded robust performance both in the test and external validation dataset. The performance was comparable to the general ophthalmologist and retinal specialist. With the extensive adoption of this technology, effective mass screening for myopic population will become feasible on a national scale.

AI-Model for Identifying Pathologic Myopia Based on Deep Learning Algorithms of Myopic Maculopathy Classification and “Plus” Lesion Detection in Fundus Images

Background: Pathologic myopia (PM) associated with myopic maculopathy (MM) and “Plus” lesions is a major cause of irreversible visual impairment worldwide. Therefore, we aimed to develop a series of deep learning algorithms and artificial intelligence (AI)–models for automatic PM identification, MM classification, and “Plus” lesion detection based on retinal fundus images.Materials and Methods: Consecutive 37,659 retinal fundus images from 32,419 patients were collected. After excluding 5,649 ungradable images, a total dataset of 32,010 color retinal fundus images was manually graded for training and cross-validation according to the META-PM classification. We also retrospectively recruited 1,000 images from 732 patients from the three other hospitals in Zhejiang Province, serving as the external validation dataset. The area under the receiver operating characteristic curve (AUC), sensitivity, specificity, accuracy, and quadratic-weighted kappa score were calculated to evaluate the classification algorithms. The precision, recall, and F1-score were calculated to evaluate the object detection algorithms. The performance of all the algorithms was compared with the experts’ performance. To better understand the algorithms and clarify the direction of optimization, misclassification and visualization heatmap analyses were performed.Results: In five-fold cross-validation, algorithm I achieved robust performance, with accuracy = 97.36% (95% CI: 0.9697, 0.9775), AUC = 0.995 (95% CI: 0.9933, 0.9967), sensitivity = 93.92% (95% CI: 0.9333, 0.9451), and specificity = 98.19% (95% CI: 0.9787, 0.9852). The macro-AUC, accuracy, and quadratic-weighted kappa were 0.979, 96.74% (95% CI: 0.963, 0.9718), and 0.988 (95% CI: 0.986, 0.990) for algorithm II. Algorithm III achieved an accuracy of 0.9703 to 0.9941 for classifying the “Plus” lesions and an F1-score of 0.6855 to 0.8890 for detecting and localizing lesions. The performance metrics in external validation dataset were comparable to those of the experts and were slightly inferior to those of cross-validation.Conclusion: Our algorithms and AI-models were confirmed to achieve robust performance in real-world conditions. The application of our algorithms and AI-models has promise for facilitating clinical diagnosis and healthcare screening for PM on a large scale.

Conbercept for Treatment of Neovascular Age-Related Macular Degeneration and Visual Impairment due to Diabetic Macular Edema or Pathologic Myopia Choroidal Neovascularization: A Systematic Review and Meta-Analysis

Background: Conbercept is a new anti-vascular endothelial growth factor (VEGF) drug. Here, we systematically conducted the efficacy, safety, compliance, and pharmacoeconomic evaluation of intravitreal conbercept (IVC) compared with other treatments in patients with neovascular age-related macular degeneration (nAMD), diabetic macular edema (DME), or pathologic myopia choroidal neovascularization (pmCNV).Methods: Databases of PubMed, Embase, Cochrane Library, ClinicalTrials.gov, SinoMed, China National Knowledge Infrastructure, and WanFang Data were systematically searched from the inception to July 27, 2021. Randomized clinical trials and pharmacoeconomic studies comparing IVC with control groups in adults with nAMD, DME, or pmCNV were reviewed and selected. Meta-analyses were performed using the fixed-effects model when pooled data were homogeneous. Heterogeneous data were analyzed using the random-effects model. Primary outcomes included visual improvement rate, mean change in visual acuity or best corrected visual acuity, and pharmacoeconomic outcomes. Additional outcomes were the mean change in fundus examination values, adverse events (AEs), quality-of-life measures, and number of injections.Results: Among 3,591 screened articles, 22 original studies with 1,910 eyes of patients were finally included. For nAMD and DME, IVC was significantly associated with better visual acuity or best corrected visual acuity improvement and fundus quantitative measures than placebo, laser photocoagulation (LP), or intravitreal triamcinolone acetonide (IVT). However, IVC showed non-inferior efficacy to intravitreal ranibizumab (IVR) according to low quality of evidence, and there was lack of trials comparing the priority of IVC to other anti-VEGF regimens. No definitive increased risk of ocular or non-ocular AEs were observed in the study groups. All patients with AEs recovered after symptomatic treatments, and no severe AEs occurred. Patients treated with IVC might have higher quality-of-life scores than those in IVR in nAMD or LP in DME. Additionally, IVC showed cost–utility advantages in nAMD and cost-effectiveness advantages than IVR in pmCNV in China.Conclusion: IVC is well-tolerated and effective for improving vision acuity and quantitative measures in fundus condition in patients with nAMD and DME compared with LP, IVT, and placebo, but gains comparable efficacy to IVR. However, well-designed, large-sample, and long-term evaluation of IVC shall be conducted in additional studies worldwide.

Deep Learning Model Based on Optical Coherence Tomography for Automated Detection of Pathologic Myopia: Prediction Model Development Study (Preprint)

BACKGROUND Pathologic myopia is a disease that causes vision impairment and blindness. Therefore, it is essential to diagnose it in a timely manner. However, there is no standardized definition for pathologic myopia, and the interpretation of pathologic myopia by optical coherence tomography is subjective and requires considerable time and money. Therefore, there is a need for a diagnostic tool that can diagnose pathologic myopia in patients automatically and in a timely manner. OBJECTIVE The purpose of this study was to develop an algorithm that uses optical coherence tomography (OCT) to automatically diagnose patients with pathologic myopia who require treatment. METHODS This study was conducted using patient data from patients who underwent optical coherence tomography tests at the Ophthalmology Department of Incheon St. Mary's Hospital and Seoul St. Mary's Hospital from January 2012 to May 2020. To automatically diagnose pathologic myopia, a deep learning model was developed using 3D optical coherence tomography images. A model was developed using transfer learning based on four pre-trained convolutional neural networks (ResNet18, ResNext50, EfficientNetB0, EfficientNetB4). The performance of each model was evaluated and compared based on accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve (AUROC). RESULTS Four models developed using test datasets were evaluated and compared. The model based on EfficientNetB4 showed the best performance (95% accuracy, 93% sensitivity, 96% specificity, and 98% AUROC). CONCLUSIONS In our study, we developed a deep learning model that can automatically diagnose pathologic myopia without segmentation of 3D optical coherence tomography images. Our deep learning model based on EfficientNetB4 demonstrated excellent performance in identifying pathologic myopia.