Associations of Peripapillary Atrophy and Fundus Tessellation with Diabetic Retinopathy

Unveiling the hidden: a deep learning approach to unraveling subzone-specific changes in peripapillary atrophy in type 2 diabetes

Purpose

This study aimed to evaluate the optical coherence tomography angiography (OCTA) changes in subzones of peripapillary atrophy (PPA) among type 2 diabetic patients (T2DM) with or without diabetic retinopathy (DR) using well-designed deep learning models.

Methods

A multi-task joint deep-learning model was trained and validated on 2,820 images to automate the determination and quantification of the microstructure and corresponding microcirculation of beta zone and gamma zone PPA. This model was then applied in the cross-sectional study encompassing 44 eyes affected by non-proliferative diabetic retinopathy (NPDR) and 46 eyes without DR (NDR). OCTA was utilized to image the peripapillary area in four layers: superficial capillary plexus (SCP), deep capillary plexus (DCP), choroidal capillary (CC) and middle-to-large choroidal vessel (MLCV).

Results

The patients in both groups were matched for age, sex, BMI, and axial length. The width and area of the gamma zone were significantly smaller in NPDR group compared to the NDR group. Multiple linear regression analysis revealed a negative association between the diagnosis of DR and the width and area of the gamma zone. The gamma zone exhibited higher SCP, DCP and MLCV density than the beta zone, while the beta zone showed higher CC density than the gamma zone. In comparison to the NDR group, the MLCV density of gamma zone was significantly lower in NPDR group, and this density was positively correlated with the width and area of the gamma zone.

Discussion

DR-induced peripapillary vascular changes primarily occur in gamma zone PPA. After eliminating the influence of axial length, our study demonstrated a negative correlation between DR and the gamma zone PPA. Longitudinal studies are required to further elucidate the role of the gamma zone in the development and progression of DR.

Prediction of the Fundus Tessellation Severity With Machine Learning Methods

Purpose

To predict the fundus tessellation (FT) severity with machine learning methods.

Methods

A population-based cross-sectional study with 3,468 individuals (mean age of 64.6 ± 9.8 years) based on Beijing Eye Study 2011. Participants underwent detailed ophthalmic examinations including fundus images. Five machine learning methods including ordinal logistic regression, ordinal probit regression, ordinal log-gamma regression, ordinal forest and neural network were used.

Main Outcome Measure

FT precision, recall, F1-score, weighted-average F1-score and AUC value.

Results

Observed from the in-sample fitting performance, the optimal model was ordinal forest, which had correct classification rate (precision) of 81.28%, while 34.75, 93.73, 70.03, and 24.82% in each classified group by FT severity. The AUC value was 0.7249. And the F1-score was 65.05%, weighted-average F1-score was 79.64% on the whole dataset. For out-of-sample prediction performance, the optimal model was ordinal logistic regression, which had precision of 77.12% on the validation dataset, while 19.57, 92.68, 64.74, and 6.76% in each classified group by FT severity. The AUC value was 0.7187. The classification accuracy of light FT group was the highest, while that of severe FT group was the lowest. And the F1-score was 54.46%, weighted-average F1-score was 74.19% on the whole dataset.

Conclusions

The ordinal forest and ordinal logistic regression model had the strong prediction in-sample and out-sample performance, respectively. The threshold ranges of the ordinal forest model for no FT and light, moderate, severe FT were [0, 0.3078], [0.3078, 0.3347], [0.3347, 0.4048], [0.4048, 1], respectively. Likewise, the threshold ranges of ordinal logistic regression model were ≤ 3.7389, [3.7389, 10.5053], [10.5053, 13.9323], > 13.9323. These results can be applied to guide clinical fundus disease screening and FT severity assessment.

Impact of age-related macular degeneration on diabetic retinopathy: An electronic health record based big data analysis from a tertiary eye centre in South India

Purpose:

To determine whether the presence of age-related macular degeneration (AMD) decreases the risk of diabetic retinopathy.

Methods:

This was a retrospective, case-cohort study performed in patients with a systemic diagnosis of diabetes at a tertiary health care center from May 2011 to April 2020. A total of 43,153 patients (1,024 AMD patients and 42,129 non-AMD patients) were included in the analysis. A total of 1,024 age and diabetes mellitus (DM) duration-matched controls were chosen from the non-AMD group for risk factor analysis. The severity of diabetic retinopathy was compared between the patients with AMD and the patients without AMD.

Results:

Out of the enrolled 43,153 diabetic patients, 26,906 were males and 16,247 were females. A total of 1,024 patients had AMD and 42,129 had no AMD. The mean age of the cohort was 58.60 ± 0.09 years. The overall prevalence of DR was noted to be 22.8% (9,825 out of 43,153 eyes). A significantly lower prevalence of diabetic retinopathy (DR) (23% in non-AMD, 11.4% in AMD, OR = −0.43, P < 0.001), non-proliferative diabetic retinopathy (NPDR) (12% in non-AMD, 8.2% in AMD, OR = −0.66, P < 0.001), and proliferative diabetic retinopathy (PDR) (11% in non-AMD, 3.2% in AMD, OR = −0.27, P < 0.001) was seen in the AMD patients. No significant difference was seen between the dry and wet AMD. On multivariate logistic regression analysis, the lower age, absence of AMD, and male gender were associated with a higher risk of PDR.

Conclusion:

The presence of AMD was noted to statistically reduce the risk of DR. Our results may be useful in the field of resource allocation and awareness of DR.

Efficacy and safety of intravitreal aflibercept for polypoidal choroidal vasculopathy: 96-week outcomes in the Japanese subgroup of the PLANET study

PURPOSE

To evaluate the efficacy and safety of intravitreal aflibercept (IVT-AFL) versus IVT-AFL plus rescue photodynamic therapy (IVT-AFL + rPDT) in the subgroup of Japanese patients with polypoidal choroidal vasculopathy (PCV) enrolled in the PLANET study.

STUDY DESIGN

A 96-week, double-masked, sham-controlled phase-3b/4 randomized clinical trial conducted at multiple centers from May 2014 to August 2016.

PATIENTS AND METHODS

Patients with PCV (BCVA 73-24 ETDRS letters [20/40-20/320 Snellen]) received 3 initial monthly doses of IVT-AFL 2 mg. At week 12, the patients were randomly assigned 1:1 to IVT-AFL + sham PDT or IVT-AFL + rPDT. Patients not requiring rescue received IVT-AFL every 8 weeks; those requiring rescue received IVT-AFL monthly plus sham/active PDT. Following week 52, the treatment intervals could be extended > 8 weeks.

RESULTS

The baseline demographics for the 159 Japanese patients were balanced. At week 96, the mean BCVA change was + 9.7 (IVT-AFL) versus + 9.5 letters (IVT-AFL + rPDT) (least-squares mean difference of - 0.3; 95% CI, - 3.7 to 3.1); the mean central subfield thickness reduction was - 148.0 µm versus - 145.9 µm. Overall, 17.1% of the patients required rescue PDT. At week 96, 25.0% (IVT-AFL) and 37.9% (IVT-AFL + rPDT) of the patients had complete polyp regression; 84.1% (IVT-AFL) and 88.4% (IVT-AFL + rPDT) of the patients had no evidence of active polyps. The mean number of injections (weeks 52-96) were 4.6 (IVT-AFL) and 4.5 (IVT-AFL + rPDT). Overall, 36.0% (IVT-AFL) and 33.8% (IVT-AFL + rPDT) of the patients experienced ocular treatment-emergent adverse events.

CONCLUSION

IVT-AFL monotherapy was efficacious for the treatment of Japanese patients with PCV, and the addition of rescue PDT did not show additional benefits.

Refractive Errors, Ocular Biometry and Diabetic Retinopathy: A Comprehensive Review

Purpose: To summarize the association between diabetic retinopathy and refractory status as well as ocular biometric parameters; To review the theories of the protective effect of high myopia against diabetic retinopathy. Methods: A comprehensive literature search on MEDLINE, EMBASE, Web of Science and Scopus databases as well as reference list search, and systematic review of relevant publications.  Results: Myopia may delay the onset and progression of diabetic retinopathy. Increased axial length in myopia is associated with reduced risk of any diabetic retinopathy and vision-threatening diabetic retinopathy. The possible mechanisms for the protective effect of myopia against diabetic retinopathy may include posterior vitreous detachment, change in retinal blood flow and oxygen demand, choroidal thinning and altered cytokine profiles.  Conclusions: High myopia may be a protective factor against the onset and progression of diabetic retinopathy. Further studies about the mechanisms of how myopia, axial length and ocular biometrics influence the onset and progression of DR are needed.

Illustration of tessellation in Down syndrome

Background: Tessellated fundus refers to a specific change in the appearance of the internal layers of the eye in which the choroidal large vessels became visible through polygonal hypopigmented areas. Such hypopigmented areas alternate with hyperpigmented zones in a tigroid pattern. Fundus tessellation is often associated with myopia and choroidal thinning.Materials and Methods: We analyzed fundus images from 50 children with Down syndrome and 52 controls.Results: Tessellation was present in 64% of children with Down syndrome, compared with only 13.5% of controls (p < .0001). In most cases, tessellation was located peripapillary, and no difference was observed in tessellation localization between children with Down syndrome and controls (p = .60). Although more prevalent in myopic children with and without Down syndrome, tessellation was present in almost half (48%) of children with Down syndrome with hyperopia versus only 5% of controls with the same refractive status.Conclusions: Mechanical stretching of the choroid could explain the high rate of tessellation in myopes. Other factors must contribute to the higher prevalence of tessellated fundus in children with Down syndrome without myopia. We discuss potentially relevant factors and propose vascular involvement as a contributor to tessellation in our population with Down syndrome. Further studies assessing choroidal vasculature in individuals with Down syndrome are needed to confirm this theory.