Spectral-Domain OCT Analysis of Risk Factors for Macular Atrophy Development in the HARBOR Study for Neovascular Age-Related Macular Degeneration.

PURPOSE: To identify baseline risk factors for macular atrophy (MA) development in HARBOR via a longitudinal assessment of monthly spectral-domain (SD)-OCT scans. Previous analyses of MA in HARBOR examined data from color fundus photography (CFP) and fluorescein angiography (FA). DESIGN: Retrospective, post hoc analysis of SD-OCT images from HARBOR (ClinicalTrials.gov identifier, NCT00891735), a phase 3, multicenter, prospective, randomized, double-blind, active treatment-controlled clinical trial. PARTICIPANTS: Patients (N = 1097) with subfoveal choroidal neovascularization secondary to neovascular age-related macular degeneration (nAMD) treated with intravitreal ranibizumab 0.5 mg monthly (n = 275), 0.5 mg pro re nata (PRN) after 3 loading doses (n = 275), 2.0 mg monthly (n = 274), or 2.0 mg PRN (n = 273). METHODS: Evaluable SD-OCT macular cube scans from patients with 24 months of follow-up (N = 941) were examined monthly from baseline to month 24 by masked reading center-trained graders. Atrophy diagnosis criteria were consistent with those proposed by the Classification of Atrophy Meetings (CAM) group: hypertransmission of light into the choroid, loss of retinal pigment epithelium, and loss of outer retinal layers. Multivariable proportional hazards regression was performed for time to atrophy development. MAIN OUTCOME MEASURES: Risk factors for MA as determined by time to MA development over 24 months of treatment. RESULTS: Baseline risk factors for MA were confirmed from prior analyses that used CFP and FA data: absence of subretinal fluid, presence of intraretinal cysts, presence of Type 3 neovascularization, and presence of atrophy in the fellow eye. This analysis of SD-OCT data identified new baseline risk factors for MA: higher central drusen volume, lower choroidal thickness, presence of nascent atrophy, presence of reticular pseudodrusen, and increased central foveal thickness. Ranibizumab treatment regimen and dose level were not found to be risk factors for MA development. CONCLUSIONS: In this analysis of a major nAMD trial using CAM atrophy criteria, new baseline risk factors for MA development were identified using an SD-OCT dataset. Risk factors for MA development identified by prior analyses were confirmed. Monthly treatment with ranibizumab 0.5 mg was not found to be a risk factor for MA development over 24 months.

Spectral-Domain OCT-Based Prevalence and Progression of Macular Atrophy in the HARBOR Study for Neovascular Age-Related Macular Degeneration.

PURPOSE: Previous studies of macular atrophy (MA) in HARBOR analyzed color fundus photography and fluorescein angiography image data. This study performed a longitudinal assessment of monthly spectral-domain (SD) OCT scans to determine MA prevalence, incidence, and progression in HARBOR. DESIGN: Post hoc analysis of SD OCT images from HARBOR (ClincalTrials.gov identifier, NCT00891735), a phase 3 multicenter, prospective, randomized, double-blind, active treatment-controlled clinical trial. PARTICIPANTS: Patients (n = 1097) with subfoveal choroidal neovascularization secondary to neovascular age-related macular degeneration (nAMD) treated with ranibizumab 0.5 mg monthly (n = 275), 0.5 mg pro re nata (PRN) after 3 loading doses (n = 275), 2.0 mg monthly (n = 274), or 2.0 mg PRN (n = 273). METHODS: Evaluable SD OCT macular cube scans from patients with 24 months of follow-up (N = 941) were examined by masked reading center-trained graders monthly from baseline to month 24. Atrophy diagnosis criteria were consistent with those proposed by the Classification of Atrophy Meetings (CAM) group: hypertransmission of light into the choroid, retinal pigment epithelium loss, and loss of outer retinal layers. Macular atrophy was considered Definite if all 3 criteria were met and Questionable if 2 were met. Study arms were compared for time to MA detection (log-rank test) and enlargement rates (time × arm interaction test). MAIN OUTCOME MEASURES: Prevalence, incidence, and enlargement rates of MA. RESULTS: At baseline, imbalance in MA rates across ranibizumab arms was evident (0.5 mg monthly, 19.1%; 0.5 mg PRN, 16.1%; 2.0 mg monthly, 10.1%; 2.0 mg PRN, 10.5%). At month 24, new MA development rates in eyes without baseline MA were similar between ranibizumab doses (0.5 mg, 25.9%; 2.0 mg, 25.4%) and treatment regimens (monthly, 26.4%; PRN, 25.0%). No significant differences in enlargement rate of new atrophy area (P = 0.479, square-root transformed) or time to detection of new MA (P = 0.997) were evident among study arms. CONCLUSIONS: In this analysis of a major nAMD trial using CAM atrophy criteria, no differences were observed in incidence or progression rates of new MA among study arms, ranibizumab doses, or treatment regimens. Monthly versus PRN treatment did not influence MA incidence or progression.

Relationship between Retinal Thickness Profiles and Visual Outcomes in Young Adults Born Extremely Preterm: The EPICure@19 Study.

PURPOSE: To quantify inner and outer retinal layer thicknesses and understand their relevance to visual function among young adults born extremely preterm (EP). DESIGN: Prospective observational study with 19 years of follow-up. PARTICIPANTS: A total of 354 eyes (226 eyes of former EP infants and 128 age-matched full-term control eyes) from 177 young adults were evaluated. Among EP participants, 50% of eyes (112/226) were not previously diagnosed with neonatal retinopathy of prematurity (ROP), 38% of eyes (84) had ROP not deemed to require treatment in the neonatal period, and 13% of eyes (30) had neonatal cryotherapy or laser ablation for ROP. METHODS: Subjects underwent eye examinations including best-corrected visual acuity (BCVA) and Heidelberg Spectralis (Heidelberg Engineering, Heidelberg, Germany) macular spectral-domain (SD) OCT imaging. Retinal layers were auto-segmented and thickness profiles were computed at the fovea by the instrument software. MAIN OUTCOME MEASURES: Correlation between retinal sublayer thickness and BCVA. RESULTS: Compared with control eyes, the inner and outer retinal layers of EP eyes were significantly thicker and BCVA was significantly reduced. Retinal layer thicknesses and BCVA were similar for untreated EP eyes and those without neonatal ROP. In contrast, treated eyes had increased inner and outer retinal layer thickness and decreased vision. Inner retinal layer thickness was moderately correlated with worse BCVA (r = 0.30, P < 0.001), but outer retinal layer thickness was not (r = -0.01, P = 0.80). Multivariate regression indicated ganglion cell layer thickness was a significant independent predictor of BCVA. CONCLUSIONS: Extremely premature birth influences maturation of the fovea and visual outcomes into early adult life. Increased ganglion cell layer thickness was associated with worse BCVA. Eyes requiring neonatal treatment for ROP had associated worse BCVA at the age of 19 years.

OCT Risk Factors for Development of Late Age-Related Macular Degeneration in the Fellow Eyes of Patients Enrolled in the HARBOR Study.

PURPOSE: To evaluate the relationship between OCT features and progression to late age related-macular degeneration (AMD) in the fellow eyes of patients enrolled in the Study of Ranibizumab Administered Monthly or on an As-needed Basis in Patients With Subfoveal Neovascular AMD (HARBOR) (ClinicalTrials.gov identifier, NCT00891735). DESIGN: Post hoc analysis of a phase 3 multicenter, prospective, randomized, double-masked, active treatment-controlled clinical trial. PARTICIPANTS: Evaluable patients (n = 501) with macular neovascularization (MNV) secondary to neovascular AMD and early or intermediate AMD in the fellow eye. METHODS: Volume OCT scans from 501 fellow eyes of 501 patients with MNV were reviewed. Baseline OCT features that were assessed included intraretinal hypereflective foci (IHRF), hyporeflective foci (hRF) within drusenoid lesions (DLs), subretinal drusenoid deposits (SDDs), and drusen volume (DV) of 0.03 mm3 or more. OCT images obtained at months 6, 12, 18, and 24 were graded by masked graders for late AMD (defined as MNV, complete retinal pigment epithelium and photoreceptor atrophy [cRORA], or both). Participant demographic characteristics (age, gender, and smoke exposure) and baseline OCT features were correlated with progression to late AMD. MAIN OUTCOME MEASURES: Incidence of late AMD, hazard ratio (HR) for demographics, and OCT risk factors. RESULTS: At month 24, 33.13% of eyes (166/501) demonstrated late AMD: 20.96% (105/501) demonstrated cRORA, whereas 12.18% (61/501) demonstrated MNV. Baseline demographic factors were not associated significantly with development of late AMD, whereas significant associations were identified for all OCT features. Intraretinal hypereflective foci had an HR of 5.21 (95% confidence interval [CI], 3.29-8.26), hRF within DLs had an HR of 2.42 (95% CI, 1.74-3.38), SDD had an HR of 1.95 (95% CI, 1.34-2.82), and DV of 0.03 mm3 or more had an HR of 1.46 (95% CI, 1.03-2.07). The correlation remained significant when considering only the progression to cRORA and MNV alone, except for DV, which was not associated significantly with progression to MNV. CONCLUSIONS: We confirmed that 4 previously reported OCT risk factors were associated with progression to late AMD in the fellow eyes of patients newly diagnosed with MNV. Although outcomes of more than 2 years were not evaluated, these findings may help to identify high-risk AMD patients.

VISUAL FUNCTION AND OPTICAL COHERENCE TOMOGRAPHY ANGIOGRAPHY FEATURES IN CHILDREN BORN PRETERM.

PURPOSE: Preterm children have an increased risk of impaired vision from retinopathy, strabismus, and high refractive error. The aim of this study was to investigate the relationship between foveal parameters generated by optical coherence tomography angiography and visual function in preterm children. METHODS: Eighty eyes (32 eyes of former preterm infants and 48 age-matched full-term control eyes) were analyzed. Subjects underwent complete eye examinations including best-corrected visual acuity and retinal imaging with the Optovue XR Avanti optical coherence tomography angiography device. Foveal morphologic parameters including foveal depth, central foveal thickness, inner retinal area, and outer retinal area were measured on a central horizontal B-scan. Foveal vasculature parameters including foveal avascular zone, superficial capillary plexus-vessel density, and deep capillary plexus-vessel density were measured on optical coherence tomography angiography. RESULTS: The best-corrected visual acuity was significantly affected in preterm children compared with controls (P < 0.0001). The central foveal thickness (P < 0.0001), inner retinal area (P = 0.01), and outer retinal area (P = 0.03) were significantly increased in preterm compared with control eyes. Foveal depth (P < 0.001) and foveal avascular zone (P < 0.001) were significantly decreased in preterm compared with control eyes. The superficial capillary plexus-vessel density (P = 0.01) and deep capillary plexus-vessel density (P = 0.003) at the fovea (1 mm) were significantly increased in preterm compared with control eyes. The best-corrected visual acuity was negatively correlated with foveal depth (r = -0.42, P = 0.001) and foveal avascular zone (r = -0.53, P < 0.001), and positively correlated with central foveal thickness (r = 0.32, P = 0.01) and inner retinal area (r = 0.32, P = 0.01), indicating that worse visual acuity was associated with a smaller foveal avascular zone, shallower foveal depth, increased central foveal thickness, and larger inner retinal area. CONCLUSION: Foveal morphology and vasculature changes in preterm children were associated with impaired visual function. Further longitudinal studies are required to evaluate these changes over time.

MACULAR MICROVASCULAR NETWORKS IN HEALTHY PEDIATRIC SUBJECTS.

PURPOSE: To report optical coherence tomography angiography (OCTA) values in healthy pediatric eyes and to identify factors that may modify these values. METHODS: In this prospective observational cross-sectional study, macular OCTA images were acquired from healthy pediatric patients. Main outcome measures were 1) foveal avascular zone (FAZ) area at the level of the superficial retinal capillary plexus (SCP); 2) SCP and deep retinal capillary plexus (DCP) perfusion density (based on the area of vessels); 3) SCP and DCP vessel density (based on a map with vessels of 1-pixel width); and 4) CC perfusion density. Multiple regression analysis was performed to assess the effect of age, sex, ethnicity, refraction, and foveal macular thickness (FMT) on OCTA parameters. RESULTS: Seventy-seven eyes from 52 subjects (23 male and 29 female) were included in analysis. Mean age was 11.1 ± 3.3 years (range = 5.0-17.0 years). Twenty-nine (55.8%) subjects were white, 14 (27.0%) Hispanic, 8 (15.4%) Asian, and 1 (1.8%) African-American. Mean refraction was -0.1 ± 2.4 diopters (D) (range = -5.75 to +9.0 D). Mean FMT was 248.6 ± 18.6 µm. Larger FAZ area was significantly associated with older age (P = 0.014). Furthermore, larger FAZ area was associated with reduced FMT (P < 0.0001). Male sex was associated only with increased SCP perfusion density (P = 0.042). Increased CC perfusion density was associated with younger age (P = 0.022). CONCLUSION: We report data for pediatric OCTA parameters in healthy subjects. Several variables influence the density of macular microvascular networks, and these factors should be considered in the OCTA study of pediatric eye disorders.

RELIABILITY OF CONFOCAL WHITE-LIGHT FUNDUS IMAGING FOR MEASUREMENT OF RETINA PIGMENT EPITHELIAL ATROPHY IN AGE-RELATED MACULAR DEGENERATION.

PURPOSE: To assess the reproducibility of confocal white-light color fundus photography (C-CFP) for the measurement of retinal pigment epithelial atrophy in comparison with confocal blue-light fundus autofluorescence (FAF) imaging and flash color fundus photography (F-CFP). METHODS: In this prospective study, eyes with age-related macular degeneration associated with evidence of retinal pigment epithelial atrophy were imaged by C-CFP, F-CFP, and FAF. Intergrader reproducibility of each modality was assessed by comparison of manual measurements by two expert graders. RESULTS: The mean areas of atrophy measured by the 2 graders were 6.67 ± 6.39, 6.35 ± 6.13, and 6.07 ± 5.48 mm for FAF, C-CFP, and F-CFP, respectively. The mean differences between the 2 graders in measuring the atrophic areas were 0.52, 0.69, and 1.62 mm for the three modalities. The intraclass correlation coefficient between the 2 graders for each modality was 0.998, 0.990, and 0.961, respectively. CONCLUSION: Measurements of atrophy from C-CFP were similar to those obtained by FAF and F-CFP. The grading reproducibility for C-CFP, however, was better than that for F-CFP and approached the level of FAF imaging. The use of C-CFP as a tool for quantitatively monitoring atrophic age-related macular degeneration lesions warrants further study, particularly in the context of clinical trials.

Impact of Multiple En Face Image Averaging on Quantitative Assessment from Optical Coherence Tomography Angiography Images.

PURPOSE: To investigate the impact of multiple en face image averaging on quantitative measurements of the retinal microvasculature using optical coherence tomography angiography (OCTA). DESIGN: Prospective, observational, cross-sectional case series. PARTICIPANTS: Twenty-one healthy individuals with normal eyes. METHODS: Macular OCTA images were acquired from all participants using the Zeiss Cirrus 5000 with Angioplex OCTA software (Carl Zeiss Meditec, Dublin, CA). Nine OCTA cube scans per eye were obtained and 9 superficial retinal layer (SRL) and deep retinal layer (DRL) en face OCTA image slabs were averaged individually after registration. Quantitative parameters from the retinal microvasculature were measured on binarized and skeletonized OCTA images and compared with single OCTA images without averaging. MAIN OUTCOME MEASURES: Vessel density (VD), vessel length density (VLD), vessel diameter index (VDI), and fractal dimension (FD). RESULTS: Participants with artifact or poor image quality were excluded, leaving 18 eyes for the analysis. After averaging, qualitatively there was apparent reduction in background noise, and fragmented vessels in the images before averaging became continuous with smoother walls and showed sharper contrast in both the SRL and DRL. Binarized and skeletonized derivates of these averaged images also showed fewer line fragments and dots in nonvascular areas and more continuous vessel images than those of images without averaging. In both SRL and DRL, VD (P = 0.0010 and P = 0.0003, respectively), VLD (P < 0.0001 for both), and FD (P < 0.0001 for both) significantly decreased and VDI significantly increased after averaging (P < 0.0001 for both). CONCLUSIONS: Averaging of multiple en face OCTA images improves image quality and also significantly impacts quantitative measurements. Reducing noise that could be misinterpreted as flow and annealing discontinuous vessel segments seem to be major mechanisms by which averaging may be of benefit.

Proposal of a simple optical coherence tomography-based scoring system for progression of age-related macular degeneration.

PURPOSE: To develop a simple, clinically practical, optical coherence tomography (OCT)-based scoring system for early age-related macular degeneration (AMD) to prognosticate risk for progression to late AMD. METHODS: We retrospectively reviewed OCT images (512 × 128 macular cube, Cirrus) from 138 patients diagnosed of early AMD in at least one eye and follow-up of at least 12 months. For patients with early AMD in both eyes, only the right eye was chosen as the study eye for longitudinal assessment. Scans were graded on four SD-OCT criteria associated with disease progression in previous studies: drusen volume within a central 3-mm circle ≥0.03 mm3, intraretinal hyperreflective foci (HRF), hyporeflective foci (hRF) within a drusenoid lesion (DL), and subretinal drusenoid deposits (SDD). Each criterion was assigned one point. For risk assessment of the study eye, the baseline status of the fellow eye was also considered, and thus these four features were also assessed in the fellow eye. The number of risk factors were summed for both eyes, yielding a total score (TS) of 0 to 8 for each patient. A fellow eye with evident choroidal neovascularization (CNV) or atrophy automatically received 4 points. Scores were then grouped into four categories to facilitate comparative analysis: I. (TS of 0, 1, 2), II. (TS of 3, 4), III. (TS of 5, 6) and IV. (TS of 7, 8). Correlation of baseline category assignment with progression to late AMD (defined as the presence of atrophy or CNV on OCT) by the last follow-up visit was evaluated with logistic regression analysis. RESULTS: The rate of progression to late AMD was 39.9% (55/138). Progression rates by category (I to IV) were 0, 14.3, 47.5, and 73.3%, respectively. Logistic regression analysis showed risk of progression to late AMD was 3.0 times (95% CI: 1.2-7.9) higher for an eye assigned to category IV than for an eye in category III and 16.4 (95% CI: 4.7-58.8) times higher than for an eye in category II. CONCLUSIONS: A simple scoring system relevant to prognosis for early AMD, and practical for use in a busy clinic, can be developed using SD-OCT criteria alone.

ASSOCIATION OF DRUSEN VOLUME WITH CHOROIDAL PARAMETERS IN NONNEOVASCULAR AGE-RELATED MACULAR DEGENERATION.

PURPOSE: The choroid is thought to be relevant to the pathogenesis of nonneovascular age-related macular degeneration, but its role has not yet been fully defined. In this study, we evaluate the relationship between the extent of macular drusen and specific choroidal parameters, including thickness and intensity. METHODS: Spectral domain optical coherence tomography images were collected from two distinct, independent cohorts with nonneovascular age-related macular degeneration: Amish (53 eyes of 34 subjects) and non-Amish (40 eyes from 26 subjects). All spectral domain optical coherence tomography scans were obtained using the Cirrus HD-OCT with a 512 × 128 macular cube (6 × 6 mm) protocol. The Cirrus advanced retinal pigment epithelium analysis tool was used to automatically compute drusen volume within 3 mm (DV3) and 5 mm (DV5) circles centered on the fovea. The inner and outer borders of the choroid were manually segmented, and the mean choroidal thickness and choroidal intensity (i.e., brightness) were calculated. The choroidal intensity was normalized against the vitreous and nerve fiber layer reflectivity. The correlation between DV and these choroidal parameters was assessed using Pearson and linear regression analysis. RESULTS: A significant positive correlation was observed between normalized choroidal intensity and DV5 in the Amish (r = 0.42, P = 0.002) and non-Amish (r = 0.33, P = 0.03) cohorts. Also, DV3 showed a significant positive correlation with normalized choroidal intensity in both the groups (Amish: r = 0.30, P = 0.02; non-Amish: r = 0.32, P = 0.04). Choroidal thickness was negatively correlated with normalized choroidal intensity in both Amish (r = -0.71, P = 0.001) and non-Amish (r = -0.43, P = 0.01) groups. Normalized choroidal intensity was the most significant constant predictor of DV in both the Amish and non-Amish groups. CONCLUSION: Choroidal intensity, but not choroidal thickness, seems to be associated with drusen volume in Amish and non-Amish populations. These observations suggest that choroidal parameters beyond thickness warrant further study in the setting of age-related macular degeneration.

Choroidal vascularity index in leptochoroid: A comparative analysis between reticular pseudodrusen and high myopia.

PURPOSE: To investigate the choroidal vascularity index (CVI) in patients affected by leptochoroid. METHODS: Three distinct age-matched cohorts were collected: patients with reticular pseudodrusen (RPD) secondary to age-related macular degeneration, patients with high-myopia, and healthy controls. CVI was calculated in the subfoveal 6000 µm diameter area. RESULTS: 54 eyes (54 patients) were included (18 eyes in each cohort). No statistical differences were disclosed in terms of age between controls, RPD patients (p = 0.062), and myopic patients (p = 0.070). Total choroidal area showed a different distribution among the 3 cohorts (p < 0.001), due to the reduction of luminal and stromal choroidal area in both RPD and myopic groups in comparison to controls (p < 0.001). Interestingly, CVI showed a different distribution between the 3 cohorts (p < 0.001). In detail, RPD group showed no changes in CVI in comparison to controls (p = 1.000), whereas the myopic group showed a higher CVI in comparison to both RPD group and controls (p < 0.001 in both analyses). CONCLUSIONS: Different changes of the choroidal vascular and stromal components characterize the leptochoroid secondary to RPD eyes and high-myopic eyes. The relative greater impairment of the vascular area in RPD eyes in comparison to myopic eyes could be at the basis of the lower development of RPD in patients with high myopia.

Optimized Prediction Models from Fundus Imaging and Genetics for Late Age-Related Macular Degeneration.

Age-related macular degeneration (AMD) is a leading cause of blindness in the developed world. In this study, we compare the performance of retinal fundus images and genetic-information-based machine learning models for the prediction of late AMD. Using data from the Age-related Eye Disease Study, we built machine learning models with various combinations of genetic, socio-demographic/clinical, and retinal image data to predict late AMD using its severity and category in a single visit, in 2, 5, and 10 years. We compared their performance in sensitivity, specificity, accuracy, and unweighted kappa. The 2-year model based on retinal image and socio-demographic (S-D) parameters achieved a sensitivity of 91.34%, specificity of 84.49% while the same for genetic and S-D-parameters-based model was 79.79% and 66.84%. For the 5-year model, the retinal image and S-D-parameters-based model also outperformed the genetic and S-D parameters-based model. The two 10-year models achieved similar sensitivities of 74.24% and 75.79%, respectively, but the retinal image and S-D-parameters-based model was otherwise superior. The retinal-image-based models were not further improved by adding genetic data. Retinal imaging and S-D data can build an excellent machine learning predictor of developing late AMD over 2-5 years; the retinal imaging model appears to be the preferred prognostic tool for efficient patient management.

Choroidal Vascularity Index in Different Cohorts of Dry Age-Related Macular Degeneration.

Purpose: The purpose of this study was to investigate the choroidal luminal and interstitial stromal alterations using choroidal vascularity index (CVI) among different cohorts of dry age-related macular degeneration (dAMD) compared to healthy subjects. Methods: Four distinct cohorts were collected: three different cohorts of patients with dAMD (i.e. drusen, reticular pseudodrusen [RPD], and geographic atrophy [GA]) and an age-matched cohort of healthy subjects (controls). CVI (the ratio between the luminal choroidal area [LCA] and the total choroidal area [TCA]) was calculated in the subfoveal 1000 µm area. Results: One hundred twenty eyes (from 120 patients) were included (30 eyes in each cohort). The mean age was 76.6 ± 7.1 years. No statistical differences were disclosed in terms of age, axial length, and central macular thickness among study groups. TCA showed a different distribution among the four cohorts (P = 0.003), mainly due to the LCA changes (P = 0.001). Interestingly, CVI showed a different distribution among the four cohorts (P < 0.001). RPD showed a lower CVI in comparison to controls (P = 0.040), whereas GA showed a lower CVI in comparison to drusen, RPD, and controls (P = 0.001, P = 0.046, and P < 0.001, respectively). Conclusions: Different cohorts of dAMD are characterized by different impairments of the choroidal vascular and stromal components, reflecting different degrees of AMD severity. Translational Relevance: CVI provides insights for better understanding the pathogenesis of AMD.

OCT Angiography Study of the Macula in Patients with Diabetic Macular Edema Treated with Intravitreal Aflibercept.

Purpose: To compare foveal avascular zone density (FAZ) in the superficial (SCP) and deep (DCP) capillary plexus and vessel density (VD) in the macula in eyes with diabetic macular edema (DME) treated with intravitreal aflibercept.Methods: Patients with DME were imaged at baseline, and 30 days after the 1st, 2nd and 3rd intravitreal aflibercept injection. Images were analyzed for the FAZ area in the SCP and DCP in each visit and VD.Results: Twenty eyes were enrolled. FAZ was 0.304 ± 0.131 mm2 in the SCP and 0.738 ± 0.5836 mm2 in the DCP at baseline. SCP FAZ was not significantly different whereas, FAZ in the DCP decreased (p = .035) after treatment. VD in the center was 20.62 ± 4.31 at baseline and decreased by 8% (p = .002). Parafoveal VD remained unchanged with treatment.Conclusion: DCP ischemia may improve after aflibercept treatment. Central macular vessel density was found to decrease post-treatment, but the clinical relevance needs further investigation.

Predicting the risk of developing diabetic retinopathy using deep learning.

BACKGROUND: Diabetic retinopathy screening is instrumental to preventing blindness, but scaling up screening is challenging because of the increasing number of patients with all forms of diabetes. We aimed to create a deep-learning system to predict the risk of patients with diabetes developing diabetic retinopathy within 2 years. METHODS: We created and validated two versions of a deep-learning system to predict the development of diabetic retinopathy in patients with diabetes who had had teleretinal diabetic retinopathy screening in a primary care setting. The input for the two versions was either a set of three-field or one-field colour fundus photographs. Of the 575 431 eyes in the development set 28 899 had known outcomes, with the remaining 546 532 eyes used to augment the training process via multitask learning. Validation was done on one eye (selected at random) per patient from two datasets: an internal validation (from EyePACS, a teleretinal screening service in the USA) set of 3678 eyes with known outcomes and an external validation (from Thailand) set of 2345 eyes with known outcomes. FINDINGS: The three-field deep-learning system had an area under the receiver operating characteristic curve (AUC) of 0·79 (95% CI 0·77-0·81) in the internal validation set. Assessment of the external validation set-which contained only one-field colour fundus photographs-with the one-field deep-learning system gave an AUC of 0·70 (0·67-0·74). In the internal validation set, the AUC of available risk factors was 0·72 (0·68-0·76), which improved to 0·81 (0·77-0·84) after combining the deep-learning system with these risk factors (p<0·0001). In the external validation set, the corresponding AUC improved from 0·62 (0·58-0·66) to 0·71 (0·68-0·75; p<0·0001) following the addition of the deep-learning system to available risk factors. INTERPRETATION: The deep-learning systems predicted diabetic retinopathy development using colour fundus photographs, and the systems were independent of and more informative than available risk factors. Such a risk stratification tool might help to optimise screening intervals to reduce costs while improving vision-related outcomes. FUNDING: Google.

Retinal Vascular Changes During Pregnancy Detected With Optical Coherence Tomography Angiography.

Purpose: To evaluate retinal vascular status during pregnancy by using optical coherence tomography angiography (OCTA). Methods: Women in their third trimester of pregnancy and nonpregnant age-matched women were recruited for this prospective, case-control study. Subjects were imaged with OCTA. Main outcome measures were foveal avascular zone parameters, perfusion density (PD) percentage in the superficial retinal capillary plexus (SCP), PD percentage in the deep retinal capillary plexus (DCP), SCP vessel length density (VLD), DCP-VLD, and choriocapillaris (CC) flow voids (i.e., flow deficits in the CC). Results: Nineteen eyes of 10 pregnant subjects and 44 eyes of 27 nonpregnant control women were included. Mean ages were 36 ± 7 and 35 ± 8 years (SD), respectively (P value = 0.78). Mean gestational age of pregnant women was 33 weeks (range = 29-39, SD = 3). There was a significant reduction in the SCP-PD in the entire scan and in the nasal Early Treatment Diabetic Retinopathy Study subfield (47.9 vs. 49.7, P = 0.04 and 49.3 vs. 51.6, P = 0.03, respectively) in the pregnant cohort versus controls. There was a significant increase in the DCP-PD in the parafoveal region and in the temporal and inferior Early Treatment Diabetic Retinopathy Study subfields (58.0 vs. 55.9, P = 0.03; 57.9 vs. 55.5, P = 0.02; 58.0 vs. 55.9, P = 0.05, respectively) in the pregnant cohort. There was no significant difference in foveal avascular zone parameters, SCP-VLD, DCP-VLD, or CC flow voids between the two populations. Conclusions: This study detected retinal vasculature changes in the third trimester of pregnancy. Mean SCP-PD was significantly decreased and mean DCP-PD was significantly increased without a difference in VLD.

Interdevice comparison of retinal sensitivity assessments in a healthy population: the CenterVue MAIA and the Nidek MP-3 microperimeters.

BACKGROUND: To compare and correlate the retinal sensitivity measurements obtained with Nidek Microperimetry-3 (MP-3) and the CenterVue Macular Integrity Assessment (MAIA) microperimeters among healthy subjects. METHODS: In this prospective comparative study, 31 eyes of 23 subjects underwent complete ophthalmological examination including retinal sensitivity assessments using two microperimeters, the MP-3 (Nidek Technologies) and the MAIA (CenterVue). The mean retinal sensitivity (dB) and its corresponding luminance (asb) and contrast (log units) were analysed between the two instruments. The interdevice reproducibility and level of agreement between the sensitivity values of the devices were assessed. RESULTS: The mean retinal sensitivity (dB) measured by the MP-3 (25.02±1.06 dB, range: 20.90-26.70) was significantly (p<0.0001) lower compared with the MAIA (30.68±0.74 dB, range: 28-31.84). The luminosity levels were significantly (p<0.0001) higher with the MP3 (7.75±1.31 asb, range: 6.44-9.06) compared with the MAIA (0.92±0.14 asb, range: 0.78-1.06). The contrast sensitivity was significantly higher for the MP-3 (0.94±0.33 log units, range: 0.61-1.27) compared with the MAIA (0.23±0.03 log units, range: 0.20-0.26). Despite these absolute differences, the intraclass coefficient was 0.85 (95% CI 0.70 to 0.92) between the two devices after applying a standard correction factor to each data point (MAIA sensitivity=MP-3 sensitivity+5.65) with a mean difference between MAIA and MP-3 of 0.01. CONCLUSION: Retinal sensitivity measures higher, but luminance and contrast sensitivity measure lower for MAIA-generated values compared with the MP-3. The relationships, however, appeared fairly consistent, and application of a standard correction factor allowed the data to be inter-related, at least for normal eyes.

Reproducibility of Macular Thickness Measurements in Eyes Affected by Dry Age-Related Macular Degeneration From Two Different SD-OCT Instruments.

BACKGROUND AND OBJECTIVE: To compare macular thickness measurement algorithms of two different spectral-domain optical coherence tomography (SD-OCT) devices in eyes affected by dry age-related macular degeneration (AMD). PATIENTS AND METHODS: Patients with dry AMD and healthy volunteers from the retina clinic of the Doheny Eye Center - UCLA were imaged using two different SD-OCT devices: the RS-3000 Advance (Nidek, Padova, Italy) and the Cirrus HD-OCT (Carl Zeiss Meditec, Dublin, CA). All patients had been previously diagnosed with drusen or geographic atrophy due to AMD. The commercial instrument software was used to generate the macular retinal thickness measurements, and measurements were compared between devices. RESULTS: Eighty-five diseased eyes from 49 patients and 16 healthy control eyes from eight normal volunteers were included in this study. The macular thickness measurements generated by the two instruments in eyes with AMD differed significantly in mean retinal thickness in the foveal center subfield (257.34 µm ± 51.72 µm using the Nidek OCT vs. 238.20 µm ± 51.89 µm using the Cirrus OCT; P < .001). The mean difference in macular thickness between the two devices was 19.14 µm ± 5.84 µm for diseased eyes and 17.06 µm ± 5.28 µm in normal control eyes, and this was not statistically different between the two groups (P > .05). The macular thickness measurements in diseased eyes, as evaluated by the two different instruments, however, showed excellent correlation (r = 0.99; P < .001), with an intraclass correlation coefficient of 0.99 (95% confidence interval, 0.98-0.99). Post hoc evaluation of cases with larger differences also showed differences in foveal center selection and variabilities in boundary selection with specific pathology. CONCLUSION: Macular thickness measurements provided by the Nidek and Cirrus OCT instruments in eyes with dry AMD are highly correlated but show a consistent difference, which may allow the use of a standard correction factor to be applied to better interrelate measurements between the devices. [Ophthalmic Surg Lasers Imaging Retina. 2018;49:410-415.].

Topographic Analysis of the Choriocapillaris in Intermediate Age-related Macular Degeneration.

PURPOSE: To quantitate regional differences in the choriocapillaris (CC) of patients with intermediate age-related macular degeneration (iAMD), using swept-source optical coherence tomography angiography (SS-OCTA) imaging. DESIGN: Cross-sectional study. METHODS: Subjects were imaged with the SS-OCTA system (PLEX Elite 9000, Carl Zeiss Meditec Inc, Dublin, California, USA). The CC en face images were first compensated for the signal attenuation caused by drusen by using the structural information from the same slab. Subsequently, for each eye, 2 compensated CC en face images generated from 2 different OCTA volume scan sets were registered and averaged. The averaged CC images were then exported to ImageJ and binarized for subsequent quantitative analysis. In addition to the analysis of the whole averaged CC en face image in iAMD eyes, quantitative analysis was also performed in 3 different regions: (1) drusen region, (2) 150-µm-wide ring around the drusen border, and (3) drusen-free region. RESULTS: Thirty eyes (30 patients) with iAMD and 30 healthy eyes from 30 controls were enrolled. Compared with controls, iAMD eyes displayed a lower number of signal voids (median and interquartile range [IQR]: 2561 and 2343-2746 vs 2734 and 2558-2834; P = .013), a greater signal void average size (median, IQR: 581.9 µm2, 466.1-726.9 µm2 vs 503.8 µm2, 429.1-576.8 µm2; P = .027), and a greater total signal void area (median, IQR: 26.0%, 22.1%-29.6% vs 23.8%, 21.2%-26.4%; P = .038). In multiple regression analysis, the presence of iAMD was not significantly associated with any of the CC variables. By contrast, the drusen region area was significantly associated with CC alterations. In the evaluation of the iAMD group, both the area underneath drusen and the 150-µm-wide ring region around drusen were characterized by an increased total signal void area (P = .005 and P = .045, respectively) vs the drusen-free region. CONCLUSIONS: Intermediate AMD eyes demonstrated increased CC flow impairment, which co-localizes to the area of CC beneath and immediately surrounding drusen.