Search

Abstract Measurement of anterior chamber depth (ACD), an important marker for the screening of primary angle-closure glaucoma, requires biometry, which is not readily used. This study assessed the relationship between ACD and health check-up data findings from participants with good corrected visual acuity in Japan. Participants underwent ophthalmic, anthropometric, and hematological assessments. The mean ACD of all 3060 participants was 3.33 ± 0.34 mm [2.22–4.72 mm]. Multivariable linear regression analysis was performed to determine factors that were significantly correlated with ACD, and logistic regression analysis was performed to predict ACD

Abstract The phenotypes of RP1-related inherited retinal dystrophies (RP1-IRD), causing autosomal dominant (AD) and autosomal recessive (AR) diseases, vary depending on specific RP1 variants. A common nonsense mutation near the C-terminus, c.5797 C > T (p.Arg1933*), is associated with RP1-IRD, but the exact role of this mutation in genotype-phenotype correlation remains unclear. In this study, we retrospectively analyzed patients with RP1-IRD (N = 42) from a single center in Japan. AR RP1-IRD patients with the c.5797 C > T mutation (N = 14) mostly displayed macular dystrophy but rarely retinitis pigmentosa or cone-rod dystrophy. Conversely, AR RP1-IRD patients without the c.5797 C > T mutation, including those with other pathogenic RP1 variants, were mostly diagnosed with severe retinitis pigmentosa. Full-field electroretinograms were significantly better in patients homozygous or compound heterozygous for the c.5797 C > T mutation than in those without this mutation, corresponding to their milder phenotypes. Clinical tests also revealed a slower onset of age and a better mean deviation value with the static visual field in AR RP1-IRD patients with the c.5797 C > T mutation compared to those without. Therefore, the presence of c.5797 C > T may partly account for the phenotypic variety of RP1-IRD and may yield milder phenotypes. These findings may be useful for predicting the prognosis of RP1-IRD patients.

Purpose: Achromatopsia (ACHM) is an early-onset cone dysfunction caused by 5 genes with cone-specific functions (CNGA3, CNGB3, GNAT2, PDE6C, and PDE6H) and by ATF6, a transcription factor with ubiquitous expression. To improve the relatively low variant detection ratio in these genes in a cohort of exome-sequenced Japanese patients with inherited retinal diseases (IRD), we performed genome sequencing to detect structural variants and intronic variants in patients with ACHM. Methods: Genome sequencing of 10 ACHM pedigrees was performed after exome sequencing. Structural, non-coding, and coding variants were filtered based on segregation between the affected and unaffected in each pedigree. Variant frequency and predicted damage scores were considered in identifying pathogenic variants. Results: A homozygous deletion involving exon 18 of RPGRIP1 was detected in 5 of 10 ACHM probands, and variant inheritance from each parent was confirmed. This deletion was relatively frequent (minor allele frequency = 0.0023) in the Japanese population but was only homozygous in patients with ACHM among the 199 Japanese IRD probands analyzed by the same genome sequencing pipeline. Conclusion: The deletion involving exon 18 of RPGRIP1 is a prevalent cause of ACHM in Japanese patients and contributes to the wide spectrum of RPGRIP1-associated IRD phenotypes, from Leber congenital amaurosis to ACHM.

Purpose To determine the factors significantly associated with the amplitudes and implicit times of the flicker electroretinograms (ERGs) recorded with the RETeval system by analyzing the comprehensive data obtained during a health checkup screening. Methods Flicker ERGs were recorded with the RETeval system from 373 individuals who had a normal fundus and optical coherence tomography images. The sex, age, anthropometric, ophthalmologic, and hematologic data were collected from all participants who were 40- to 89-years-of-age. Univariable and multivariable linear mixed effects regression analyses were performed to identify factors that were significantly associated with the implicit times and amplitudes of the RETeval flicker ERGs. Results Univariable linear mixed effects regression analysis showed significant correlations between the implicit times and the best-corrected visual acuity, the age, the axial length, the blood sugar level, and the blood urea nitrogen level. Analyses by multivariable linear mixed effects regression identified that the axial length (β = 0.28), the age (β = 0.24), and the blood sugar level (β = 0.092) were three independent factors that were significantly correlated with the implicit times of the RETeval flicker ERGs. Univariable linear mixed effects regression analysis also showed significant correlations between the amplitudes of the RETeval flicker ERGs and the age, the platelet count, and the creatinine level. Multivariable linear mixed effects regression models identified the age (β = -0.092), the platelet count (β = 0.099), and the creatinine level (β = -0.12) as three independent factors that were significantly correlated with the amplitudes of the RETeval flicker ERGs. However, the smoking habits, body mass index, and the blood pressure were not significantly correlated with either the implicit times or amplitudes of the RETeval flicker ERGs. Conclusions Our results indicate that the age and some ophthalmologic and hematologic findings but not the anthropometric findings were significantly associated with the implicit times and amplitudes of the RETeval flicker ERGs. Thus, clinicians should remember these factors when analyzing the RETeval flicker ERGs.

Eye checkups have become increasingly important to maintain good vision and quality of life. As the population requiring eye checkups increases, so does the clinical work burden of clinicians. An automatic screening algorithm to reduce the clinicians’ workload is necessary. Machine learning (ML) has recently become one of the chief techniques for automated image recognition and is a helpful tool for identifying ocular diseases. However, the accuracy of ML models is lower in a clinical setting than in the laboratory. The performance of ML models depends on the training dataset. Eye checkups often prioritize speed and minimize image processing. Data distribution differs from the training dataset and, consequently, decreases prediction performance. The study aim was to investigate an ML model to screen for retinal diseases from low-quality optical coherence tomography (OCT) images captured during actual eye chechups to prevent a dataset shift. The ensemble model with convolutional neural networks (CNNs) and random forest models showed high screening performance in the single-shot OCT images captured during the actual eye checkups. Our study indicates the strong potential of the ensemble model combining the CNN and random forest models in accurately predicting abnormalities during eye checkups.