PurposeTo investigate the relationship between baseline number of hyperreflective foci (HF) on spectral domain optical coherence tomography (SD-OCT) in patients with diabetic macular edema (DME), as well as the dynamics of HF during treatment with anti-vascular endothelial growth factor (VEGF), and treatment response.MethodsWe evaluated patients diagnosed with DME scheduled for treatment with intravitreal bevacizumab. Eyes were classified as adequate or insufficient treatment responders based on logMAR visual acuity improvement and central retinal thickness (CRT) decrease after three consecutive injections. Associations between number of HF at baseline and treatment response, the change in HF over the course of treatment, and the distribution of HF within the retinal layers were evaluated.ResultsIn 54 eyes of 41 patients, mean number of HF and CRT decreased after intravitreal treatment with bevacizumab (p = 0.002 and p<0.001 respectively). Decrease in CRT after 3 months was independently associated with a higher number of HF at baseline (estimated effect -2.61, 95% CI [-4.42–-0.31], p = 0.006). Eyes with adequate treatment response presented with more HF at baseline (OR 1.106, 95% CI [1.012–1.210], p = 0.030) than eyes with insufficient treatment response. Most HF were located within the inner retinal layers, and decrease of HF was mostly due to a decrease of inner retinal HF.ConclusionsIn patients with DME treated with anti-VEGF, higher baseline numbers of HF have predictive value for treatment response in terms of visual acuity improvement and CRT decrease after 3 months. In addition, HF were responsive to anti-VEGF therapy.

We developed a deep learning algorithm for the automatic segmentation and quantification of intraretinal cystoid fluid (IRC) in spectral domain optical coherence tomography (SD-OCT) volumes independent of the device used for acquisition. A cascade of neural networks was introduced to include prior information on the retinal anatomy, boosting performance significantly. The proposed algorithm approached human performance reaching an overall Dice coefficient of 0.754 ± 0.136 and an intraclass correlation coefficient of 0.936, for the task of IRC segmentation and quantification, respectively. The proposed method allows for fast quantitative IRC volume measurements that can be used to improve patient care, reduce costs, and allow fast and reliable analysis in large population studies.

Retinal microaneurysms in DME could be classified topographically and morphologically by OCTA. OCTA detected less microaneurysms than FFA, and this appeared to be dependent on leakage activity and retinal thickening. Morphological appearance of microaneurysms (irregular, fusiform and mixed saccular/fusiform) was associated with increased leakage activity and retinal thickening.

We propose a method for automatic detection of the foveal center in optical coherence tomography (OCT). The method is based on a pixel-wise classification of all pixels in an OCT volume using a fully convolutional neural network (CNN) with dilated convolution filters. The CNN-architecture contains anisotropic dilated filters and a shortcut connection and has been trained using a dynamic training procedure where the network identifies its own relevant training samples. The performance of the proposed method is evaluated on a data set of 400 OCT scans of patients affected by age-related macular degeneration (AMD) at different severity levels. For 391 scans (97.75%) the method identified the foveal center with a distance to a human reference less than 750 μm, with a mean (± SD) distance of 71 μm ± 107 μm. Two independent observers also annotated the foveal center, with a mean distance to the reference of 57 μm ± 84 μm and 56 μm ± 80 μm, respectively. Furthermore, we evaluate variations to the proposed network architecture and training procedure, providing insight in the characteristics that led to the demonstrated performance of the proposed method.

Background/AimTo validate a previously developed model for prediction of diabetic retinopathy (DR) for personalised retinopathy screening in persons with type 1 diabetes.MethodsRetrospective medical data of persons with type 1 diabetes treated in an academic hospital setting were used for analysis. Sight-threatening retinopathy (STR) was defined as the presence of severe non-proliferative DR, proliferative DR or macular oedema. The presence and grade of retinopathy, onset of diabetes, systolic blood pressure, and levels of haemoglobin A1c were used to calculate an individual risk estimate and personalised screening interval. In persons with STR, the occurrence was compared with the calculated date of screening. The model’s predictive performance was measured using calibration and discrimination techniques.ResultsOf the 268 persons included in our study, 24 (9.0%) developed STR during a mean follow-up of 4.6 years. All incidences of STR occurred after the calculated screening date. By applying the model, the mean calculated screening interval was 30.5 months, which is a reduction in screening frequency of 61% compared with annual screening and 21% compared with biennial screening. The discriminatory ability was good (Harrell’s C-statistic=0.82, 95% CI 0.74 to 0.90), and calibration showed an overestimation of risk in persons who were assigned to a higher risk for STR.ConclusionThis validation study suggests that a screening programme based on the previously developed prediction model is safe and efficient. The use of a personalised screening frequency could improve cost-effectiveness of diabetic eye care.

This study focuses on further characterization of the audiovestibular phenotype and on genotype-phenotype correlations of DFNB77, an autosomal recessive type of hearing impairment (HI). DFNB77 is associated with disease-causing variants in LOXHD1, and is genetically and phenotypically highly heterogeneous. Heterozygous deleterious missense variants in LOXHD1 have been associated with late-onset Fuchs corneal dystrophy (FCD). However, up to now screening for FCD of heterozygous carriers in DFNB77 families has not been reported. This study describes the genotype and audiovestibular phenotype of 9 families with DFNB77. In addition, carriers within the families were screened for FCD. Fifteen pathogenic missense and truncating variants were identified, of which 12 were novel. The hearing phenotype showed high inter- and intrafamilial variation in severity and progression. There was no evidence for involvement of the vestibular system. None of the carriers showed (pre-clinical) symptoms of FCD. Our findings expand the genotypic and phenotypic spectrum of DFNB77, but a clear correlation between the type or location of the variant and the severity or progression of HI could not be established. We hypothesize that environmental factors or genetic modifiers are responsible for phenotypic differences. No association was found between heterozygous LOXHD1 variants and the occurrence of FCD in carriers.

PurposeTo investigate risk factors for the development and progression of diabetic retinopathy (DR) and long‐term visual outcomes in Dutch patients with type 1 diabetes mellitus (T1DM).MethodsCumulative incidences were calculated for DR, vision‐threatening DR (VTDR), defined as (pre)proliferative DR and diabetic macular oedema, and best‐corrected visual acuity (BCVA) <0.5 and <0.3 at the most recent eye examination. The following factors were assessed: duration of diabetes, age of onset of T1DM, gender, mean HbA1c, HbA1c variability (defined as coefficient of variation of five separate HbA1c measurements), mean arterial blood pressure, body mass index, albuminuria and lipid profile. We used multivariable Cox regression models to identify factors associated with DR development and progression to VTDR.ResultsWe found 25‐year cumulative incidences of 63% for DR, 21% for VTDR, 2% for BCVA <0.5, and 1% for BCVA <0.3. Mean HbA1c (HR 1.023, p < 0.001), HbA1c variability (HR 1.054, p < 0.001), age of onset of T1DM (HR 1.024, p < 0.001), HDL cholesterol (HR 0.502, p = 0.002) and total cholesterol (HR 1.210, p = 0.029) showed an independent association with faster development of any form of DR. The mean HbA1c (HR 1.023, p < 0.001) and the presence of albuminuria (HR 2.940, p = 0.028) were associated with faster progression to VTDR.ConclusionThese data show relatively low cumulative incidences of DR, VTDR and visual impairment. Higher mean HbA1c, HbA1c variability, age of onset of T1DM and total cholesterol were independently associated with the risk of DR development, and a protective association was found for HDL cholesterol in subjects with T1DM. Mean HbA1c and presence of albuminuria were associated with progression of DR.

Ophthalmological abnormalities in facioscapulohumeral dystrophy may lead to treatable vision loss, facilitate diagnostics, could help unravelling the pathophysiology and serve as biomarkers. In this study, we provide a detailed description of the ophthalmological findings in a well-defined cohort of patients with facioscapulohumeral dystrophy using state of the art retina imaging techniques. Thirty-three genetically confirmed patients (aged 7–80 years) and 24 unrelated healthy controls (aged 6–68 years) underwent clinical ophthalmological examination, fundus photography, optical coherence tomography/angiography, genotyping and neurological examination. All patients had normal corrected visual acuity and normal intraocular pressure. In 27 of the 33 patients, weakness of the orbicularis oculi was observed. Central retinal pathology, only seen in patients and not in healthy controls, included twisting (tortuosity) of the retinal arteries in 25 of the 33 patients and retinal pigment epithelium defects in 4 of the 33 patients. Asymmetrical foveal hypoplasia was present in three patients, and exudative abnormalities were observed in one patient. There was a correlation between the severity of retinal tortuosity and the D4Z4 repeat array size (R2 = 0.44, P < 0.005). Follow-up examination in a subgroup of six patients did not show any changes after 2 years. To conclude, retinal abnormalities were frequent but almost always subclinical in patients with facioscapulohumeral dystrophy and consisted primarily of arterial tortuosity and foveal abnormalities. Retinal tortuosity was seen in the retinal arterioles and correlated with the D4Z4 repeat array size, thereby providing clinical evidence for an underlying genetic linkage between the retina and facioscapulohumeral dystrophy.