Lloyd Paul Aiello, MD, PhD

OBJECTIVE To assess self-reported awareness of diabetic retinopathy (DR) and concordance of eye examination follow-up compared with findings from concurrent retinal images. RESEARCH DESIGN AND METHODS We conducted a prospective observational 10-year study of 26,876 consecutive patients with diabetes who underwent retinal imaging during an endocrinology visit. Awareness and concordance were evaluated using questionnaires and retinal imaging. RESULTS Awareness information and gradable images were available in 25,360 patients (94.3%). Severity of DR by imaging was as follows: no DR (n = 14,317; 56.5%), mild DR (n = 6,805; 26.8%), or vision-threatening DR (vtDR; n = 4,238; 16.7%). In the no, mild, and vtDR groups, 96.7%, 88.5%, and 54.9% of patients, respectively, reported being unaware of any prior DR. When DR was …

ImportanceMachine learning (ML) algorithms have the potential to identify eyes with early diabetic retinopathy (DR) at increased risk for disease progression.ObjectiveTo create and validate automated ML models (autoML) for DR progression from ultra-widefield (UWF) retinal images.Design, Setting and ParticipantsDeidentified UWF images with mild or moderate nonproliferative DR (NPDR) with 3 years of longitudinal follow-up retinal imaging or evidence of progression within 3 years were used to develop automated ML models for predicting DR progression in UWF images. All images were collected from a tertiary diabetes-specific medical center retinal image dataset. Data were collected from July to September 2022.ExposureAutomated ML models were generated from baseline on-axis 200° UWF retinal images. Baseline retinal images were labeled for progression based on centralized reading center …

OBJECTIVE To investigate quantitative and qualitative changes in retinal structure using optical coherence tomography (OCT) and their associations with systemic or other risk factors in individuals with type 1 diabetes (T1D). RESEARCH DESIGN AND METHODS In the Epidemiology of Diabetes Interventions and Complications (EDIC) study, OCT images were obtained during study years 25–28 (2019–2022) in 937 participants; 54% and 46% were from the original intensive (INT) and conventional (CONV) glycemic management treatment groups, respectively. RESULTS Average age for participants was 61 years old, diabetes duration 39 years, and HbA1c 7.6%. Participants originally in the CONV group were more likely to have disorganization of retinal inner layers (DRIL) (CONV 27.3% vs. INT 18.7%; P = 0.0003), intraretinal fluid (CONV 24.4% vs. INT 19.2 …

TOPICTo review preclinical and clinical evidence on basic and cellular mechanisms potentially pertinent to diabetic retinal disease (DRD) that might eventually be relevant to update the DRD staging system.CLINICAL RELEVANCEHyperglycemia is a major risk factor for early lesions of DRD. Updating the DRD staging system to incorporate relevant basic and cellular mechanisms pertinent to DRD is necessary to better address early disease, disease progression, the use of therapeutic interventions, and treatment effectiveness.METHODSThe Basic and Cellular Mechanisms Working Group (BCM-WG) of the Mary Tyler Moore Vision Initiative carefully and extensively reviewed available preclinical and clinical evidence through multiple iterations and classified these into evidence grids, level of supporting evidence and anticipated future relevance to DRD.RESULTSA total of 40 identified targets based on …

PurposeTo compare Early Treatment Diabetic Retinopathy Study (ETDRS) diabetic retinopathy (DR) severity on ultrawide field (UWF) colour imaging (CI) and UWF fluorescein angiography (FA).DesignCross-sectional retrospective review.SubjectsPatients with diabetes mellitus and at least mild non-proliferative DR on UWF-CI.MethodsUWF-CI and UWF-FA images acquired within 1 month of each other were evaluated independently using ETDRS DR Severity Scale (DRSS) for colour photography adapted for UWF-CI and UWF-FA. Extent of non-perfusion (NP, mm2) was determined from UWF-FA images.Main outcome measuresAgreement rate between DRSS on UWF-CI and UWF-FA.ResultsImages from 218 eyes of 137 patients with diabetes were evaluated. Agreement rate for DRSS between UWF-CI and UWF-FA was moderate to substantial (K=0.46, Kw=0.65). Over-all, DRSS was worse in 73 (33.5%) eyes …

Purpose: To create and validate automated deep learning models (autoML) for diabetic retinopathy (DR) progression from ultrawide field (UWF) retinal images.Methods: This was a prospective development and validation study of autoML models for predicting DR progression in UWF images. A total of 1,179 unique de-identified UWF images with mild or moderate nonproliferative DR (NPDR) from 703 persons with diabetes and 3 years of longitudinal follow-up retinal imaging had AutoML models generated from baseline on-axis 200-degree UWF retinal images. Baseline retinal images were graded for progression based on centralized reading center evaluation of baseline and follow-up images according to the clinical Early Treatment Diabetic Retinopathy Study severity scale. Images for model development were split 8-1-1 for training, optimization, and testing to detect DR progression of 1 or more steps. Algorithm validation was performed using a 328 image set from the same patient population but not used in model development. The outcomes measured were area under the precision-recall curve (AUPRC), sensitivity, specificity, positive predictive value, negative predictive value,(SN, SP, PPV, NPV, respectively) accuracy, and F1 scores.Results: DR severity distribution was 32.2% mild and 67.8% moderate NPDR. DR progression was present in 50% of the training set. The model’s AUPRC for baseline mild NPDR was 0.717 and 0.863 for moderate NPDR. Performance on the validation set for mild NPDR was 0.70, 0.72, 0.31, 0.93, and 0.15 (SN, SP, PPV, NPV, and prevalence) and for moderate NPDR was 0.82, 0.74, 0.38, 0.93, and 0.22. In …

Although diabetic retinopathy (DR) and age-related macular degeneration (AMD) are the two most common sight-threatening conditions in the elderly population, our understanding of the associated risks for these diseases have not been studied. Thus, we examined the association between AMD and DR in people with type 1 and type 2 diabetes from the Joslin 50-Year Medalist Study, composed of individuals who have had insulin-dependent diabetes for 50 years or longer and Beetham Eye Institute. In Medalists, there was an inverse association between DR severity and the presence of AMD (N= 1163, P=< 0.0001, AMD: 23.4%, 18.6, 7.9% for no-mild, moderate-severe, and proliferative DR, respectively). Similarly, the presence of AMD was associated with milder DR in the BEI cohort (N= 58, P= 0.004). There were no eyes with neovascular (severe AMD). The presence of AMD was associated with older age (P= 0 …

OBJECTIVE To determine whether individuals with type 1 diabetes (T1D) who develop any retinopathy at any time prior to 5 years of diabetes duration have an increased subsequent risk for further progression of retinopathy or onset of proliferative diabetic retinopathy (PDR), clinically significant macular edema (CSME), diabetes-related retinal photocoagulation, or anti-vascular endothelial growth factor injections. Additionally, to determine the influence of HbA1c and other risk factors in these individuals. RESEARCH DESIGN AND METHODS Diabetic retinopathy (DR) was assessed longitudinally using standardized stereoscopic seven-field fundus photography at time intervals of 6 months to 4 years. Early-onset DR (EDR) was defined as onset prior to 5 years of T1D duration. Cox models assessed the associations of EDR with subsequent risk of outcomes …

Purpose: To compare the performance of automated machine learning (autoML) models to detect diabetic retinopathy (DR) using a diverse versus non-diverse training dataset of retinal images.Methods: A previously validated AutoML (Google Cloud) model to detect DR was generated based on 5-field handheld retinal images (N= 16,681) from a Filipino (Malay/Asian, FIL) population. Additional images using the same camera from an Egyptian, Finnish, and Tanzanian (EGY, FIN, TANZ) cohort were tested against this model (N= 641,767,215). The baseline algorithm trained with Filipino eyes was re-trained with additional eyes from different ethnicities (Arab, White, Black). The performance of the augmented algorithm was retested on a holdout set from the same 3 ethnicities. Image labeling was performed at a reading center using the International DR and diabetic macular edema (DME) scale. Referable DR (refDR) was defined as moderate nonproliferative DR or worse, or DME.Results: RefDR per ethnicity (FIL, EGY, FIN, TZA,%): 17.3/87.8/18.5/27.9. Baseline performance of the model trained on Filipino eyes to detect refDR per ethnicity: 0.995/0.838/0.931/0.856 [area under the precision-recall (AUPRC), Fig. 1]. For all non-Asian eyes combined (N= 1,623), the baseline Filipino model had an AUPRC of 0.901. After augmenting the baseline Filipino model with non-Asian eyes, AUPRC for detection of refDR in the whole cohort increased to 0.957. Number of images for validation of the model per ethnicity: 225/164/192/49. Sensitivity/specificity (SN/SP) of the baseline Filipino model tested against the FIL, EGY, FIN, TZA images: 0.96/0.98, 0.77/0.90, 0 …

Purpose: To evaluate the association between the presence of age-related macular degeneration (AMD) and diabetic retinopathy severity (DR).Methods: Digital fundus images were graded for ETDRS DR severity and for AMD using the Age-Related Eye Disease Study (AREDS) classification system in people with type 1 and type 2 diabetes at the Beetham Eye Institute (BEI) of Joslin Diabetes Center, and individuals who have had insulin-dependent diabetes for 50 years or longer (Joslin 50-Year Medalist Study). Individual retinal layer thicknesses in the foveal area were assessed in eyes of Medalists (N= 545). Retinol binding protein 3 (RBP3) and vascular endothelial growth factor (VEGF) concentrations were measured by ELISA in vitreous samples (N= 187) obtained during vitreoretinal surgery at the BEI and from postmortem eyes of Medalists.Results: The presence of AMD was associated with less severe DR in both Medalist (N= 1163, P< 0.0001) and BEI cohorts (N= 58, P= 0.004). In Medalists, 16.2% had any AMD, 13.2% had AREDS level 1-2, and 3.0% AREDS level 3-4. There were no eyes with neovascular AMD. All eyes with AMD in the BEI cohort had AREDS level 1-2 (10.4%). In the Medalist cohort, the presence and severity of AMD was positively associated with the presence of drusen outside the macula (P< 0.0001). The presence of drusen outside the macula was associated with less severe DR in Medalists (P= 0.04). In a subset of Medalists (22.9%) with longitudinal follow up, DR progression was less frequent in eyes with AMD AREDS level≥ 2 (N= 2) compared to eyes with AMD AREDS level< 2 (N= 31, 13.0% vs 7.1%, P= 0.04 …

Introduction The purpose of this study was to compare 2-field (2F) and 5-field (5F) mydriatic handheld retinal imaging for the assessment of diabetic retinopathy (DR) severity in a community-based DR screening program (DRSP). Methods This was a prospective, cross-sectional diagnostic study, evaluating images of 805 eyes from 407 consecutive patients with diabetes acquired from a community-based DRSP. Mydriatic standardized 5F imaging (macula, disc, superior, inferior, temporal) with handheld retinal camera was performed. 2F (disc, macula), and 5F images were independently assessed using the International DR classification at a centralized reading center. Simple (K) and weighted (Kw) kappa statistics were calculated for DR. Sensitivity and specificity for referable DR ([refDR] moderate nonproliferative DR [NPDR] or worse) and vision-threatening DR ([vtDR] severe NPDR or worse) for 2F compared to …

ImportanceAnti–vascular endothelial growth factor (VEGF) injections in eyes with nonproliferative diabetic retinopathy (NPDR) without center-involved diabetic macular edema (CI-DME) reduce development of vision-threatening complications from diabetes over at least 2 years, but whether this treatment has a longer-term benefit on visual acuity is unknown.ObjectiveTo compare the primary 4-year outcomes of visual acuity and rates of vision-threatening complications in eyes with moderate to severe NPDR treated with intravitreal aflibercept compared with sham. The primary 2-year analysis of this study has been reported.Design, Setting, and ParticipantsRandomized clinical trial conducted at 64 clinical sites in the US and Canada from January 2016 to March 2018, enrolling 328 adults (399 eyes) with moderate to severe NPDR (Early Treatment Diabetic Retinopathy Study [ETDRS] severity level 43-53; range, 0 …

Purpose: To determine the ability of a semi-automated UWF directed OCT imaging algorithm to detect sub-clinical neovascularization elsewhere (NVE) in eyes with severe NPDR on UWF color imaging.Methods: Consecutive patients with severe NPDR who presented to a retina clinic were imaged using a novel semi-automated algorithm to detect NVE. The algorithm acquires 7 consecutive 120-line scans (figure 1: central 9x12 mm raster and 6 mid-peripheral 6x6 mm scans (infero and supero-nasal, infero and supero-temporal, inferior and superior). A trained grader (MA) evaluated each individual line scan to determine the presence of NVE. NVE were graded to be present if the lesion clearly demonstrated a breach in the internal limiting membrane (ILM) on at least 2 consecutive line scans. An entire field considered ungradable if 50% of the line scans were absent or of poor quality limiting the ability to determine retinal morphology or identify the ILM.Results: A total of 14 eyes of 13 patients were included in this study, 38.5% were female. Mean age was 45.6±12.7 years, 46.2% type 1, mean duration of diabetes was 18.6±8.9 years and mean HbA1c of 8.6±1.7%. In this cohort, 7/14 (50%) were found to have NVE previously undetected by UWF color imaging (a single NVE in 5 eyes and 2 NVEs in 2 eyes). These NVEs were detected on Raster scan (3), inferior scans (2), supero-temporal scans (2), nasal scans (1) and infero-nasal scans (1). When looking at consecutive line scans with detected NVE, the mean number of lines with NVE was 10.8±8.1 scans (range 2-28). For individual scan areas the overall ungradable rate was 14.3% and by field …

Purpose: Glycemic control and diabetes duration are the strongest predictors of the progression of diabetic retinopathy (DR). However, some individuals, despite the long duration of diabetes, do not develop DR or show only mild nonproliferative DR (NPDR). These “extreme” phenotype patients are unlikely to experience further worsening of DR. In this study, using a well-defined phenotypic strategy, we seek to understand better the role of genetic variants that may protect people from developing advanced DR.Methods: We explored two distinct patient cohorts: Group 1, the" No DR/Very Mild NPDR" cohort despite the long duration of diabetes (> 20 years; n= 104), and Group 2, the" Proliferative Diabetic Retinopathy" cohort group (n= 170). Ultrawide field retinal images were graded for DR severity at a masked central reading center by ETDRS standard grading protocol. We leveraged the whole genome and exome to define the" protective" genetic factors from the cohort. To test for association between rare or common variants and No DR phenotype, we performed multimarker SNP-set Kernel Association Test (SKAT) analysis. A combined functional prediction (CFP) algorithm was used to evaluate the impact of coding variants in the cohort.

The Mary Tyler Moore Vision Initiative Diabetic Retinal Disease (DRD) Clinical Endpoints Workshop was held on October 22, 2022 to accelerate progress toward establishment of useful clinical and research endpoints and development of new therapeutics that have important relevance across the full spectrum of DRD pathology. More than 90 patient representatives, clinicians, scientists, funding and regulatory agencies, diagnostic, therapeutic and biotech industry representatives discussed the needs for new diagnostic and therapeutic approaches to prevent and restore retinal neurovascular unit integrity. Phase I of the MTM Vision Initiative plans, notably updating the DRD staging system and severity scale, establishing a human ocular biorepository and resource, and clinical endpoints and biomarker development and validation, was emphasized.

Purpose: To evaluate the ability of ultrawide field (UWF)–directed optical coherence tomography (OCT) to detect retinal neovascularization in eyes thought to have severe nonproliferative diabetic retinopathy (NPDR).Methods: Retrospective study of 20 consecutive patients diagnosed with severe NPDR by clinical examination. All patients underwent UWF color imaging (UWF-CI) and UWF-directed OCT following a prespecified imaging protocol to assess the mid periphery, 15/32 (46.9%) eyes underwent UWF–fluorescein angiography (FA). On OCT, new vessels elsewhere (NVE) were defined when vessels breached the internal limiting membrane.Results: A total of 32 eyes of 20 patients were evaluated. Of the 45 suspected areas of intraretinal microvascular abnormalities (IRMA) on UWF-CI, 38 (84.4%) were imaged by UWF-directed OCT, and 9/38 IRMA (23.7%) were NVE by OCT. Furthermore, UWF-directed OCT identified seven additional NVE in three eyes not seen on UWF-CI. This resulted in a change in diabetic retinopathy (DR) severity from severe NPDR to PDR in 8/32 eyes (25.0%). Among the 46.9% of eyes with UWF-FA, UWF-directed OCT agreed with the UWF-FA findings in 80%(12/15 eyes), missing only one peripheral NVE outside the UWF-OCT scanning area. Two eyes had subtle NVD that were not evident on UWF-directed OCT.Conclusions: This pilot study suggests that UWF-directed OCT may help differentiate IRMA from NVE and detect unrecognized NVE in eyes with advanced DR in a clinical practice setting. Future prospective studies in larger cohorts could determine whether this rapid and noninvasive method is clinically …

Purpose: Factor XII (FXII) and its activated serine protease form FXIIa have been identified in the vitreous of patients with diabetic macular edema (DME). FXIIa is the primary activator of the kallikrein-kinin system, which has been implicated in DME. The current study investigates the effects of FXII on retinal edema and neuroretinal responses in mice.Methods: Age-matched male wild-type (WT) and FXII deficient (FXIIKO) mice received intravitreal injections (1µL) of VEGF (100ng/eye), FXIIa (50ng/eye), or saline control. Retinal thickness was measured at 24 and 48 hours post injection using spectral domain optical coherence tomography. Retinal neuronal function was assessed using full field dark-adapted scotopic electroretinography (ERG) at 48 hours following intravitreal injections. Diabetes was induced in WT and FXIIKO mice by streptozotocin and scotopic ERG was measured after 4 months duration of diabetes …

Purpose: The clinical utility of 200 colour ultra-wide field imaging using a red (635 nm) and green (532 nm) composite image (optomap) has been demonstrated in over 2000 peer-reviewed publications in a multitude of retinal conditions.[1-4]. This study was designed to determine whether the addition of a blue reflectance image (488nm) to the existing RG composite image generated by the Optos California P200DTx Scanning Laser Ophthalmoscope (SLO) provides a non-inferior image for retinal evaluation.Methods: A prototype RGB device was used to collect a variety of images of diseased and normal retinas for evaluation. The composite RG image will remain unchanged on this product. Subjects were imaged under an IRB approved protocol in Houston, Texas on both the RGB prototype and on commercially-available California device.

PurposeTo create and validate code-free automated deep learning models (AutoML) for diabetic retinopathy (DR) classification from handheld retinal images.DesignProspective development and validation of AutoML models for DR image classification.ParticipantsA total of 17 829 deidentified retinal images from 3566 eyes with diabetes, acquired using handheld retinal cameras in a community-based DR screening program.MethodsAutoML models were generated based on previously acquired 5-field (macula-centered, disc-centered, superior, inferior, and temporal macula) handheld retinal images. Each individual image was labeled using the International DR and diabetic macular edema (DME) Classification Scale by 4 certified graders at a centralized reading center under oversight by a senior retina specialist. Images for model development were split 8-1-1 for training, optimization, and testing to detect …

OBJECTIVE To describe the relationships between the cumulative incidences of long-term complications in individuals with type 1 diabetes (T1D) and assess whether observed associations are independent of age, duration of diabetes, and glycemic levels. METHODS Proliferative diabetic retinopathy (PDR), clinically significant macular edema (CSME), reduced estimated glomerular filtration rate (eGFR), amputations, cardiovascular disease (CVD), and mortality were assessed in the Diabetes Control and Complications Trial/Epidemiology of Diabetes Interventions and Complications Study over ∼30 years. RESEARCH DESIGN AND RESULTS The cumulative incidence of complications ranged from 3% (amputations) to 37% (CSME). There were large differences in the cumulative incidence of PDR between participants with versus without prior CSME (66% vs …