ABSTRACT
BACKGROUND: Continuous-flow left ventricular assist device (cfLVAD) use is effective in supporting patients with end-stage heart failure (ESHF). Reduced flow pulsatility within the systemic circulation in cfLVAD-supported patients may lead to alterations within the microcirculation. Temporal changes in microvasculature in relation to adverse events in cfLVAD-supported patients have not been studied. We aimed to profile changes within retinal microvasculature and its association with adverse events. METHODS: Retinal photography was performed using Topcon TRC-NW8 nonmydriatic fundus camera in cfLVAD-supported patients and ESHF control patients. Specific retinal measurements were evaluated using a validated semiautomated program. Demographic and adverse event data were documented. RESULTS: Forty-eight patients were studied (n = 29 cfLVAD, n = 19 ESHF). There were significant trends in retinal arteriolar caliber (B = -0.53 µm, 95% confidence interval [CI]: -0.96 to -0.10, p = 0.016) and retinal fractal dimension parameters (B = 0.014, 95% CI: 0.001-0.002, p = 0.016) in linear mixed model regressions. Among cfLVAD patients, there was a significant association between the incidence of gastrointestinal bleeding and stepwise increases in retinal arteriolar-venular caliber ratio (hazard ratio: 3.03, 95% CI: 2.06-4.45, p = 0.005), a measure of arteriolar narrowing. CONCLUSIONS: We have observed for the first time that alterations in retinal microvasculature in cfLVAD-supported patients may be associated with gastrointestinal bleeding. While understanding these temporal changes may predict future adverse events in cfLVAD-supported patients, further multicenter studies are required to confirm the associations observed.
ABSTRACT
AIMS: To develop and externally test deep learning (DL) models for assessing the image quality of three-dimensional (3D) macular scans from Cirrus and Spectralis optical coherence tomography devices. METHODS: We retrospectively collected two data sets including 2277 Cirrus 3D scans and 1557 Spectralis 3D scans, respectively, for training (70%), fine-tuning (10%) and internal validation (20%) from electronic medical and research records at The Chinese University of Hong Kong Eye Centre and the Hong Kong Eye Hospital. Scans with various eye diseases (eg, diabetic macular oedema, age-related macular degeneration, polypoidal choroidal vasculopathy and pathological myopia), and scans of normal eyes from adults and children were included. Two graders labelled each 3D scan as gradable or ungradable, according to standardised criteria. We used a 3D version of the residual network (ResNet)-18 for Cirrus 3D scans and a multiple-instance learning pipline with ResNet-18 for Spectralis 3D scans. Two deep learning (DL) models were further tested via three unseen Cirrus data sets from Singapore and five unseen Spectralis data sets from India, Australia and Hong Kong, respectively. RESULTS: In the internal validation, the models achieved the area under curves (AUCs) of 0.930 (0.885-0.976) and 0.906 (0.863-0.948) for assessing the Cirrus 3D scans and Spectralis 3D scans, respectively. In the external testing, the models showed robust performance with AUCs ranging from 0.832 (0.730-0.934) to 0.930 (0.906-0.953) and 0.891 (0.836-0.945) to 0.962 (0.918-1.000), respectively. CONCLUSIONS: Our models could be used for filtering out ungradable 3D scans and further incorporated with a disease-detection DL model, allowing a fully automated eye disease detection workflow.
ABSTRACT
INTRODUCTION: Vision and hearing impairments are highly prevalent and have a significant impact on physical, psychological and social wellbeing. There is a need for accurate, contemporary national data on the prevalence, risk factors and impacts of vision and hearing loss in Australian adults. OBJECTIVES: The Australian Eye and Ear Health Survey (AEEHS) aims to determine the prevalence, risk factors and impacts of vision and hearing loss in both Aboriginal and Torres Strait Islander and non-Indigenous older adults. METHODS AND ANALYSIS: The AEEHS is a population-based cross-sectional survey which will include 5,000 participants (3250 non-Indigenous aged 50 years or older and 1750 Aboriginal and Torres Strait Islander people aged 40 years or older) from 30 sites covering urban and rural/regional geographic areas, selected using a multi-stage, random cluster sampling strategy. Questionnaires will be administered to collect data on socio-demographic, medical, ocular and ontological history. The testing battery includes assessment of blood pressure, blood sugar, anthropometry, visual acuity (presenting, unaided, pinhole and best-corrected), refraction, tonometry, slit lamp and dilated eye examination, ocular imaging including optical coherence tomography (OCT), OCT-angiography and retinal photography, and automated visual fields. Audiometry, tympanometry and video otoscopy will also be performed. The primary outcomes are age-standardised prevalence of cause-specific vision and hearing impairment. Secondary outcomes are prevalence of non-blinding eye diseases (including dry eye disease), patterns in health service utilisation, universal health coverage metrics, risk factors for vision and hearing impairment, and impact on quality of life.
Subject(s)
Health Surveys , Hearing Loss , Adult , Aged , Female , Humans , Male , Middle Aged , Australia/epidemiology , Cross-Sectional Studies , Hearing Loss/epidemiology , Prevalence , Risk Factors , Vision Disorders/epidemiology , Australian Aboriginal and Torres Strait Islander PeoplesABSTRACT
BACKGROUND/OBJECTIVES: To assess the prevalence and correlates of impaired activities of daily living (ADLs) in patients with neovascular age-related macular degeneration (nAMD) who present for anti-vascular endothelial growth factor (VEGF) therapy. METHODS: In a clinic-based cohort of 437 patients with nAMD who presented for anti-VEGF therapy, the Older American Resources and Services Scale (OARS) was administered to assess for impairments in basic, instrumental and total ADL. Logistic regression analyses were conducted to determine odds ratios (OR) and 95% confidence intervals (CI) for factors associated with ADL impairment. RESULTS: The prevalence of impaired basic, instrumental and total ADL was 37.76%, 67.82% and 39.59%, respectively. In multivariate-adjusted models, moderate visual impairment [OR 5.65, 95% CI (2.31-13.83) and blindness [OR 5.43, 95% CI (2.09-14.12)] were associated with greater odds of impaired total ADL. Depressive symptoms [OR 2.08, 95% CI (1.08-4.00)], the presence of any disability [OR 3.16, 95% CI (1.64-0.07)] and never driving [OR 4.00, 95% CI (1.60-10.00)] were also positively associated with total ADL impairment. Better vision-related quality of life (QoL) was inversely associated with impaired instrumental ADL whilst higher health-related QoL scores were associated with decreased odds of total ADL impairment. CONCLUSIONS: There is a high prevalence rate of ADL impairment among nAMD patients presenting for therapy. Visual impairment, never driving, poor physical and mental health increased the odds of experiencing ADL impairment whilst better VRQoL and HRQoL reduced the odds of impairment.
Subject(s)
Activities of Daily Living , Angiogenesis Inhibitors , Vascular Endothelial Growth Factor A , Visual Acuity , Humans , Male , Female , Aged , Prevalence , Angiogenesis Inhibitors/therapeutic use , Risk Factors , Vascular Endothelial Growth Factor A/antagonists & inhibitors , Aged, 80 and over , Visual Acuity/physiology , Wet Macular Degeneration/drug therapy , Wet Macular Degeneration/epidemiology , Wet Macular Degeneration/physiopathology , Quality of Life , Ranibizumab/administration & dosage , Ranibizumab/therapeutic use , Intravitreal Injections , Middle Aged , Cross-Sectional StudiesSubject(s)
Diabetes Mellitus , Diabetic Retinopathy , Humans , Diabetic Retinopathy/complications , Diabetic Retinopathy/diagnosis , Diabetic Retinopathy/epidemiology , Vitreous Hemorrhage/diagnosis , Vitreous Hemorrhage/epidemiology , Vitreous Hemorrhage/etiology , Incidence , Vitreous Body , VitrectomyABSTRACT
CLINICAL RELEVANCE: Valid and updated clinical indicators can serve as important tools in assessing and improving eyecare delivery. BACKGROUND: Indicators for diabetic eyecare in Australia were previously developed from guidelines published before 2013 and then used to assess the appropriateness of care delivery through a nationwide patient record card audit (the iCareTrack study). To reflect emerging evidence and contemporary practice, this study aimed to update clinical indicators for optometric care for people with type 2 diabetes in Australia. METHODS: Forty-five candidate indicators, including existing iCareTrack and new indicators derived from nine high-quality evidence-based guidelines, were generated. A two-round modified Delphi process where expert panel members rated the impact, acceptability, and feasibility of the indicators on a 9-point scale and voted for inclusion or exclusion of the candidate indicators was used. Consensus on inclusion was reached when the median scores for impact, acceptability, and feasibility were ≥7 and >75% of experts voted for inclusion. RESULTS: Thirty-two clinical indicators with high acceptability, impact and feasibility ratings (all median scores: 9) were developed. The final indicators were related to history taking (n = 12), physical examination (n = 8), recall period (n = 5), referral (n = 5), and patient education/communication (n = 2). Most (14 of 15) iCareTrack indicators were retained either in the original format or with modifications. New indicators included documenting the type of diabetes, serum lipid level, pregnancy, systemic medications, nephropathy, Indigenous status, general practitioner details, pupil examination, intraocular pressure, optical coherence tomography, diabetic retinopathy grading, recall period for high-risk diabetic patients without retinopathy, referral of high-risk proliferative retinopathy, communication with the general practitioner, and patient education. CONCLUSION: A set of 32 updated diabetic eyecare clinical indicators was developed based on contemporary evidence and expert consensus. These updated indicators inform the development of programs to assess and enhance the eyecare delivery for people with diabetes in Australia.
ABSTRACT
OBJECTIVE: To develop a deep learning algorithm (DLA) to detect diabetic kideny disease (DKD) from retinal photographs of patients with diabetes, and evaluate performance in multiethnic populations. MATERIALS AND METHODS: We trained 3 models: (1) image-only; (2) risk factor (RF)-only multivariable logistic regression (LR) model adjusted for age, sex, ethnicity, diabetes duration, HbA1c, systolic blood pressure; (3) hybrid multivariable LR model combining RF data and standardized z-scores from image-only model. Data from Singapore Integrated Diabetic Retinopathy Program (SiDRP) were used to develop (6066 participants with diabetes, primary-care-based) and internally validate (5-fold cross-validation) the models. External testing on 2 independent datasets: (1) Singapore Epidemiology of Eye Diseases (SEED) study (1885 participants with diabetes, population-based); (2) Singapore Macroangiopathy and Microvascular Reactivity in Type 2 Diabetes (SMART2D) (439 participants with diabetes, cross-sectional) in Singapore. Supplementary external testing on 2 Caucasian cohorts: (3) Australian Eye and Heart Study (AHES) (460 participants with diabetes, cross-sectional) and (4) Northern Ireland Cohort for the Longitudinal Study of Ageing (NICOLA) (265 participants with diabetes, cross-sectional). RESULTS: In SiDRP validation, area under the curve (AUC) was 0.826(95% CI 0.818-0.833) for image-only, 0.847(0.840-0.854) for RF-only, and 0.866(0.859-0.872) for hybrid. Estimates with SEED were 0.764(0.743-0.785) for image-only, 0.802(0.783-0.822) for RF-only, and 0.828(0.810-0.846) for hybrid. In SMART2D, AUC was 0.726(0.686-0.765) for image-only, 0.701(0.660-0.741) in RF-only, 0.761(0.724-0.797) for hybrid. DISCUSSION AND CONCLUSION: There is potential for DLA using retinal images as a screening adjunct for DKD among individuals with diabetes. This can value-add to existing DLA systems which diagnose diabetic retinopathy from retinal images, facilitating primary screening for DKD.
Subject(s)
Deep Learning , Diabetes Mellitus, Type 2 , Diabetic Nephropathies , Diabetic Retinopathy , Humans , Diabetic Retinopathy/diagnosis , Diabetes Mellitus, Type 2/complications , Cross-Sectional Studies , Longitudinal Studies , Australia , AlgorithmsABSTRACT
INTRODUCTION: The aim of this study was to determine the prevalence of diabetic retinopathy (DR) in a low socioeconomic region of a high-income country, as well as determine the diagnostic utility of point-of-care screening for high-risk populations in tertiary care settings. RESEARCH DESIGN AND METHODS: This was a cross-sectional study of patients with diabetes attending foot ulcer or integrated care diabetes clinics at two Western Sydney hospitals (n=273). DR was assessed using portable, two-field, non-mydriatic fundus photography and combined electroretinogram/ pupillometry (ERG). With mydriatic photographs used as the reference standard, sensitivity and specificity of the devices were determined. Prevalence of DR and vision-threatening diabetic retinopathy (VTDR) were reported, with multivariate logistic regression used to identify predictors of DR. RESULTS: Among 273 patients, 39.6% had any DR, while 15.8% had VTDR, of whom 59.3% and 62.8% were previously undiagnosed, respectively. Non-mydriatic photography demonstrated 20.2% sensitivity and 99.5% specificity for any DR, with a 56.7% screening failure rate. Meanwhile, mydriatic photography produced high-quality images with a 7.6% failure rate. ERG demonstrated 72.5% sensitivity and 70.1% specificity, with a 15.0% failure rate. The RETeval ERG was noted to have an optimal DR cut-off score at 22. Multivariate logistic regression identified an eGFR of ≤29 mL/min/1.73 m2, HbA1c of ≥7.0%, pupil size of <4 mm diameter, diabetes duration of 5-24 years and RETeval score of ≥22 as strong predictors of DR. CONCLUSION: There is a high prevalence of vision-threatening and undiagnosed DR among patients attending high-risk tertiary clinics in Western Sydney. Point-of-care DR screening using portable, mydriatic photography demonstrates potential as a model of care which is easily accessible, targeted for high-risk populations and substantially enhances DR detection.
Subject(s)
Diabetes Mellitus , Diabetic Retinopathy , Humans , Diabetic Retinopathy/diagnosis , Diabetic Retinopathy/epidemiology , Point-of-Care Systems , Cross-Sectional Studies , Mass Screening/methods , Sensitivity and Specificity , MydriaticsABSTRACT
OBJECTIVES: To report the 15-year incidence of driving cessation and its associated vision-related risk factors in an older Australian population-based cohort. STUDY DESIGN: 15-year data from a sample of 2379 participants who indicated that they were driving at baseline from The Blue Mountains Eye Study was analysed. Questions about driving cessation was asked at all four visits and was recorded as a binary response (Yes/No). Clinical vision examinations were performed at each visit to determine presenting and best-corrected visual acuity and any incident eye diseases (Yes/No). MAIN OUTCOME MEASURES: The cumulative 15-year incidence of driving cessation was calculated using interval-censored data progression-free survival analyses. Age- and sex-adjusted and multivariable-adjusted interval-censored Cox proportional hazard models were used to report the hazard ratios (HRs) for associations of baseline and incident vision status with driving cessation. RESULTS: The 15-year cumulative incidence of driving cessation amongst the 2379 participants was 20.7 %, with women more likely to cease driving than men (p = 0.0005). Cataract (HR 1.98 (95 % confidence interval(Cl) 1.45-2.71)) and age-related macular degeneration (HR 1.85 (95%Cl 1.37-2.50)) were associated with increased risk of driving cessation whilst presenting and best-corrected visual acuity in the better eye were protective against cessation (presenting: HR 0.96 (95%Cl 0.95-0.98); best-corrected: HR 0.93 (95%Cl 0.91-0.95)) in age- and sex-adjusted models, with these factors remaining independently associated in the multivariable-adjusted models. CONCLUSION: Cumulative incidence of driving cessation increased with older age and was higher in females. Cataract and age-related macular degeneration were independently associated with cessation, whilst better visual acuity at baseline helped prolong driving.
Subject(s)
Cataract , Macular Degeneration , Male , Humans , Female , Incidence , Australia , Visual Acuity , Risk FactorsABSTRACT
Background: The cardiovascular risk associated with different levels of hypertensive retinopathy, including mild, remains unclear. We performed an individual participant meta-analysis from 6 population-based cohort studies to determine the relationship of hypertensive retinopathy with incident cardiovascular outcomes. Methods: We identified cohort studies that objectively assessed hypertensive retinopathy from photographs, documented incident cardiovascular outcomes, and were population-based. Six studies contributed data from 11,013 individuals at baseline with 5-13 years follow-up. Participants were recruited if they had hypertension and did not have confounding conditions such as diabetic retinopathy. Main outcome measures were incident coronary heart disease (CHD), stroke and a composite endpoint of cardiovascular disease (CHD or stroke). Pooled estimates of incident risk ratios (IRR) were obtained after adjusting for age, gender, systolic blood pressure, serum total cholesterol, high density lipoprotein and smoking. Results: Among eligible participants with hypertension and without diabetes, there were 1018/9662 (10.5%) incident CHD events, 708/11,013 (6.4%) incident stroke events and 1317/9378 (14.0%) incident CVD events. Mild hypertensive retinopathy was associated with increased risk of CVD (IRR 1.13, 95% CI 1.00 to 1.27) and CHD (IRR 1.17, 95% CI 1.02 to 1.34) but not stroke; moderate hypertensive retinopathy was associated with increased risk of CVD (IRR 1.25 95% CI 1.02 to 1.53) but not stroke or CHD individually. Conclusions: In persons with hypertension, both mild and moderate hypertensive retinopathy were associated with higher CVD risk.
ABSTRACT
Neovascular age-related macular degeneration (nAMD) is a major cause of visual impairment and blindness. Anti-vascular endothelial growth factor (VEGF) agents, such as ranibizumab, bevacizumab, aflibercept, brolucizumab and faricimab have revolutionized the clinical management of nAMD. However, there remains an unmet clinical need for new and improved therapies for nAMD, since many patients do not respond optimally, may lose response over time or exhibit sub-optimal durability, impacting on real world effectiveness. Evidence is emerging that targeting VEGF-A alone, as most agents have done until recently, may be insufficient and agents that target multiple pathways (e.g., aflibercept, faricimab and others in development) may be more efficacious. This article reviews issues and limitations that have arisen from the use of existing anti-VEGF agents, and argues that the future may lie in multi-targeted therapies including alternative agents and modalities that target both the VEGF ligand/receptor system as well as other pathways.
Subject(s)
Angiogenesis Inhibitors , Macular Degeneration , Humans , Angiogenesis Inhibitors/therapeutic use , Ranibizumab/therapeutic use , Bevacizumab/therapeutic use , Recombinant Fusion Proteins/therapeutic use , Intravitreal InjectionsABSTRACT
OBJECTIVE: There are limited data on the influence of ethnicity on diabetic retinopathy (DR). We sought to determine the distribution of DR by ethnic group in Australia. DESIGN: Clinic-based cross-sectional study. SETTING: Participants with diabetes in a defined geographical region of Sydney, Australia, who attended a tertiary retina referral clinic. PARTICIPANTS: The study recruited 968 participants. INTERVENTION: Participants underwent a medical interview and retinal photography and scanning. PRIMARY OUTCOME MEASURES: DR was defined from two-field retinal photographs. Diabetic macular oedema (DMO) was defined from spectral domain optical coherence tomography (OCT-DMO). The main outcomes were any DR, proliferative DR (PDR), clinically significant macular oedema (CSME), OCT-DMO and sight-threatening DR (STDR). RESULTS: There was high proportion of any DR (52.3%), PDR (6.3%), CSME (19.7%), OCT-DMO (28.9%) and STDR (31.5%) in people attending a tertiary retinal clinic. Participants of Oceanian ethnicity had the highest proportion of any DR and STDR (70.4% and 48.1%, respectively), while the lowest proportion was in participants of East Asian ethnicity (38.3% and 15.8%, respectively). Proportion of any DR and STDR in Europeans was 54.5% and 30.3%, respectively. Independent predictive factors for diabetic eye disease were ethnicity, longer duration of diabetes, higher glycated haemoglobin and higher blood pressure. Even after adjusting for risk factors, Oceanian ethnicity remained associated with twofold higher odds of any DR (adjusted OR 2.10, 95% CI 1.10 to 4.00) and all other forms of DR including STDR (adjusted OR 2.22, 95% CI 1.19 to 4.15). CONCLUSION: In people attending a tertiary retinal clinic, the proportion of people with DR varies among ethnic groups. The high proportion in persons of Oceanian ethnicity suggests a need for targeted screening of this at-risk group. In addition to traditional risks factors, ethnicity may be an additional independent predictor of DR.
Subject(s)
Diabetes Mellitus, Type 2 , Diabetic Retinopathy , Macular Edema , Humans , Diabetic Retinopathy/diagnosis , Ethnicity , Macular Edema/etiology , Cross-Sectional Studies , Retina , Diabetes Mellitus, Type 2/complicationsABSTRACT
BACKGROUND/OBJECTIVES: To examine the risk factors for poor vision-related and health-related quality of life (QoL) in patients with neovascular age-related macular degeneration (nAMD) who present for anti-vascular endothelial growth factor (anti-VEGF) therapy. METHODS: In a clinic-based cohort of 547 nAMD patients who presented for treatment, the National Eye Institute Visual Function Questionnaire-25 (NEI-VFQ25), Short-Form 36 (SF-36) and EuroQoL EQ-5D-5L questionnaires were administered to assess vision-related and health-related QoL. Of these, 83 participants were followed up one-year later to provide longitudinal data. RESULTS: Individuals with mild or moderate visual impairment or blindness at baseline had significantly lower NEI-VFQ-25 scores at follow-up. The presence of ≥3 chronic diseases was associated with lower SF-36 mental component scores (MCS) (p = 0.04) and EQ-VAS scores (p = 0.05). Depressive symptoms were associated with significantly lower MCS (p < 0.0001) and EQ-VAS scores (p = 0.02). Individuals with versus without impaired basic activities of daily living (ADLs) exhibited NEI-VFQ-25 and EQ-VAS scores that were 10.96 (p = 0.03) and 0.13 (p = 0.02) points lower. Those with impaired instrumental ADLs scored 11.62 (p = 0.02), 13.13 (p < 0.0001) and 15.8 (p = 0.0012) points lower in the NEI-VFQ-25, SF-36 physical component score and EQ-5D-5L summary score, respectively. CONCLUSIONS: The QoL of nAMD patients is affected by visual acuity as well as patients' medical history, mental health and functional status.
Subject(s)
Macular Degeneration , Quality of Life , Humans , Quality of Life/psychology , Activities of Daily Living , Vision, Ocular , Surveys and Questionnaires , Risk Factors , Sickness Impact ProfileABSTRACT
STUDY OBJECTIVES: There has been long-standing interest in potential links between obstructive sleep apnea (OSA) and eye disease. This study used retinal photography to identify undiagnosed retinal abnormalities in a cohort of sleep clinic patients referred for polysomnography (PSG) and then determined associations with PSG-quantified sleep-disordered breathing (SDB) severity. METHODS: Retinal photographs (n = 396 patients) were taken of each eye prior to polysomnography and graded according to validated, standardized, grading scales. SDB was quantified via in-laboratory polysomnography (PSG; n = 385) using standard metrics. A questionnaire (n = 259) documented patient-identified pre-existing eye disease. Within-group prevalence rates were calculated on a per patient basis. Data were analyzed using multivariate logistic regression models to determine independent predictors for retinal abnormalities. P < 0.05 was considered significant. RESULTS: Main findings were (1) 76% of patients reported no pre-existing "eye problems"; (2) however, 93% of patients had at least one undiagnosed retinal photograph-identified abnormality; (3) most common abnormalities were drusen (72%) and peripapillary atrophy (PPA; 47%); (4) age was the most common risk factor; (5) diabetes history was an expected risk factor for retinopathy; (6) patients with very severe levels of SDB (apnea hypopnea index ≥ 50 events/h) were nearly three times more likely to have PPA. CONCLUSION: Retinal photography in sleep clinic settings will likely detect a range of undiagnosed retinal abnormalities, most related to patient demographics and comorbidities and, except for PPA, not associated with SDB. PPA may be indicative of glaucoma, and any association with severe SDB should be confirmed in larger prospective studies.
Subject(s)
Sleep Apnea Syndromes , Sleep Apnea, Obstructive , Humans , Polysomnography , Prospective Studies , Sleep Apnea Syndromes/diagnosis , Sleep Apnea Syndromes/epidemiology , Sleep Apnea Syndromes/complications , Sleep , Sleep Apnea, Obstructive/diagnosis , Sleep Apnea, Obstructive/epidemiology , Sleep Apnea, Obstructive/complicationsABSTRACT
Aims: Computer-aided detection systems for retinal fluid could be beneficial for disease monitoring and management by chronic age-related macular degeneration (AMD) and diabetic retinopathy (DR) patients, to assist in disease prevention via early detection before the disease progresses to a "wet AMD" pathology or diabetic macular edema (DME), requiring treatment. We propose a proof-of-concept AI-based app to help predict fluid via a "fluid score", prevent fluid progression, and provide personalized, serial monitoring, in the context of predictive, preventive, and personalized medicine (PPPM) for patients at risk of retinal fluid complications. Methods: The app comprises a convolutional neural network-Vision Transformer (CNN-ViT)-based segmentation deep learning (DL) network, trained on a small dataset of 100 training images (augmented to 992 images) from the Singapore Epidemiology of Eye Diseases (SEED) study, together with a CNN-based classification network trained on 8497 images, that can detect fluid vs. non-fluid optical coherence tomography (OCT) scans. Both networks are validated on external datasets. Results: Internal testing for our segmentation network produced an IoU score of 83.0% (95% CI = 76.7-89.3%) and a DICE score of 90.4% (86.3-94.4%); for external testing, we obtained an IoU score of 66.7% (63.5-70.0%) and a DICE score of 78.7% (76.0-81.4%). Internal testing of our classification network produced an area under the receiver operating characteristics curve (AUC) of 99.18%, and a Youden index threshold of 0.3806; for external testing, we obtained an AUC of 94.55%, and an accuracy of 94.98% and an F1 score of 85.73% with Youden index. Conclusion: We have developed an AI-based app with an alternative transformer-based segmentation algorithm that could potentially be applied in the clinic with a PPPM approach for serial monitoring, and could allow for the generation of retrospective data to research into the varied use of treatments for AMD and DR. The modular system of our app can be scaled to add more iterative features based on user feedback for more efficient monitoring. Further study and scaling up of the algorithm dataset could potentially boost its usability in a real-world clinical setting. Supplementary information: The online version contains supplementary material available at 10.1007/s13167-022-00301-5.
ABSTRACT
OBJECTIVE: Cardiac autonomic neuropathy (CAN) may contribute to vascular complications in diabetes. We hypothesized that adolescents with CAN are at greater risk of diabetic retinopathy and early kidney dysfunction. RESEARCH DESIGN AND METHODS: In this prospective longitudinal study of 725 adolescents with type 1 diabetes without retinopathy and albuminuria at baseline, early CAN was defined as one or more abnormalities in seven heart rate tests derived from a 10-min electrocardiogram. Retinopathy was defined as the presence of one or more microaneurysms, early kidney dysfunction as an albumin excretion rate (AER) >7.5 µg/min, and albuminuria as an AER >20 µg/min. Multivariable generalized estimating equations were used to examine the association between CAN and retinopathy or early kidney dysfunction. Cox proportional hazards regression analysis was used to assess cumulative risks of incident retinopathy and albuminuria. RESULTS: At baseline, the mean age of the sample was 13.6 ± 2.6 years, 52% were male, and mean diabetes duration was 6.1 ± 3.3 years. Over a median follow-up of 3.8 (interquartile range 2.2-7.5) years, the complication rate 27% for retinopathy, 16% for early kidney dysfunction, and 3% for albuminuria. The mean study HbA1c was 72.3 ± 16 mmol/mmol (8.6 ± 1.4%). CAN predicted incident retinopathy (odds ratio 2.0 [95% CI 1.4, 2.9]) and early kidney dysfunction (1.4 [1.0, 2.0]) after adjusting for HbA1c and diabetes duration. CAN also predicted retinopathy (hazard ratio 1.57 [95% CI 1.09, 2.26]) and albuminuria (2.30 [1.05, 5.04]) independently of HbA1c. CONCLUSIONS: CAN predicted incident retinopathy and kidney dysfunction in adolescents with type 1 diabetes, likely reflecting autonomic microvascular dysregulation contributing to complications. Therefore, screening and interventions to reduce CAN may influence the risk of complications.
Subject(s)
Diabetes Mellitus, Type 1 , Diabetic Nephropathies , Diabetic Retinopathy , Adolescent , Albumins , Albuminuria/complications , Albuminuria/etiology , Autonomic Pathways , Child , Diabetes Mellitus, Type 1/complications , Diabetic Retinopathy/complications , Diabetic Retinopathy/etiology , Female , Glycated Hemoglobin , Humans , Kidney , Longitudinal Studies , Male , Prospective Studies , Risk FactorsABSTRACT
OBJECTIVE: Current guidelines recommend biennial diabetic retinopathy (DR) screening commencing at the age of 11 years and after 2-5 years' duration of type 1 diabetes. Growing evidence suggests less frequent screening may be feasible. RESEARCH DESIGN AND METHODS: Prospective data were collected from 2,063 youth with type 1 diabetes who were screened two or more times between 1990 and 2019. Baseline (mean ± SD) age was 13.3 ± 1.8 years, HbA1c was 8.6 ± 1.3% (70.1 ± 14.7 mmol/mol), diabetes duration was 5.6 ± 2.8 years, and follow-up time was 4.8 ± 2.8 years. DR was manually graded from 7-field retinal photographs using the Early Treatment Diabetic Retinopathy Study (ETDRS) scale. Markov chain was used to calculate probabilities of DR change over time and hazard ratio (HR) of DR stage transition. RESULTS: The incidence of moderate nonproliferative DR (MNPDR) or worse was 8.6 per 1,000 patient-years. Probabilities of transition to this state after a 3-year interval were from no DR, 1.3%; from minimal DR, 5.1%; and from mild DR, 22.2%, respectively. HRs (95% CIs) for transition per 1% current HbA1c increase were 1.23 (1.16-1.31) from no DR to minimal NPDR, 1.12 (1.03-1.23) from minimal to mild NPDR, and 1.28 (1.13-1.46) from mild to MNPDR or worse. HbA1c alone explained 27% of the transitions between no retinopathy and MNPDR or worse. The addition of diabetes duration into the model increased this value to 31% (P = 0.03). Risk was also increased by female sex and higher attained age. CONCLUSIONS: These results support less frequent DR screening in youth with type 1 diabetes without DR and short duration. Although DR progression to advanced stages is generally slow, higher HbA1c greatly accelerates it.
Subject(s)
Diabetes Mellitus, Type 1 , Diabetes Mellitus, Type 2 , Diabetic Retinopathy , Adolescent , Child , Diabetes Mellitus, Type 1/complications , Diabetes Mellitus, Type 2/complications , Diabetic Retinopathy/diagnosis , Diabetic Retinopathy/epidemiology , Diabetic Retinopathy/etiology , Female , Glycated Hemoglobin , Humans , Male , Prospective Studies , Risk FactorsABSTRACT
OBJECTIVES: To appraise the existing literature reporting an association between retinal markers and cognitive impairment in adults aged 65 years and over and to provide directions for future use of retinal scanning as a potential tool for dementia diagnosis. DESIGN: Systematic review of peer-reviewed empirical articles investigating the association of retinal markers in assessing cognitive impairment. DATA SOURCES: Three electronic databases, Medline, PsycINFO and EMBASE were searched from inception until March 2022. ELIGIBILITY CRITERIA: All empirical articles in English investigating the association between retinal markers and cognition in humans aged ≥65 years using various retinal scanning methodologies were included. Studies with no explicit evaluation of retinal scanning and cognitive outcomes were excluded. Risk of bias was assessed using the Quality Assessment of Diagnostic Accuracy Studies tool. DATA EXTRACTION AND SYNTHESIS: Data extraction was conducted by two authors (VJ, RS) and reviewed by another author (JS). Results were synthesised and described narratively. RESULTS: Sixty-seven eligible studies examining 6815 older adults were included. Majority of studies were cross-sectional (n=60; 89.6%). Optical coherence tomography (OCT) was the most commonly used retinal scanning methodology to measure the thickness of retinal nerve fibre layer, the ganglion cell complex, choroid and macula. 51.1% of cross-sectional studies using OCT reported an association between the thinning of at least one retinal parameter and poor cognition. Longitudinal studies (n=6) using OCT also mostly identified significant reductions in retinal nerve fibre layer thickness with cognitive decline. Study quality was overall moderate. CONCLUSION: Retinal nerve fibre layer thickness is linked with cognitive performance and therefore may have the potential to detect cognitive impairment in older adults. Further longitudinal studies are required to validate our synthesis and understand underlying mechanisms before recommending implementation of OCT as a dementia screening tool in clinical practice. PROSPERO REGISTRATION NUMBER: CRD42020176757.
Subject(s)
Cognitive Dysfunction , Dementia , Aged , Cognition , Cognitive Dysfunction/diagnosis , Dementia/diagnosis , Humans , Retina/diagnostic imaging , Tomography, Optical Coherence/methodsABSTRACT
OBJECTIVE: To examine trends in diabetic retinopathy (DR) and diabetic macular edema (DME) in adolescents with type 1 diabetes between 1990 and 2019. RESEARCH DESIGN AND METHODS: We analyzed 5,487 complication assessments for 2,404 adolescents (52.7% female, aged 12-20 years, diabetes duration >5 years), stratified by three decades (1990-1999, 2000-2009, 2010-2019). DR and DME were graded according to the modified Airlie House classification from seven-field stereoscopic fundal photography. RESULTS: Over three decades, the prevalence of DR was 40, 21, and 20% (P < 0.001) and DME 1.4, 0.5, and 0.9% (P = 0.13), respectively, for 1990-1999, 2000-2009, and 2010-2019. Continuous subcutaneous insulin infusion (CSII) use increased (0, 12, and 55%; P < 0.001); mean HbA1c was bimodal (8.7, 8.5, and 8.7%; P < 0.001), and the proportion of adolescents meeting target HbA1c <7% did not change significantly (8.3, 7.7, and 7.1%; P = 0.63). In multivariable generalized estimating equation analysis, DR was associated with 1-2 daily injections (odds ratio 1.88, 95% CI 1.42-2.48) and multiple injections in comparison with CSII (1.38, 1.09-1.74); older age (1.11, 1.07-1.15), higher HbA1c (1.19, 1.05-1.15), longer diabetes duration (1.15, 1.12-1.18), overweight/obesity (1.27, 1.08-1.49) and higher diastolic blood pressure SDS (1.11, 1.01-1.21). DME was associated with 1-2 daily injections (3.26, 1.72-6.19), longer diabetes duration (1.26, 1.12-1.41), higher diastolic blood pressure SDS (1.66, 1.22-2.27), higher HbA1c (1.28, 1.03-1.59), and elevated cholesterol (3.78, 1.84-7.76). CONCLUSIONS: One in five adolescents with type 1 diabetes had DR in the last decade. These findings support contemporary guidelines for lower glycemic targets, increasing CSII use, and targeting modifiable risk factors including blood pressure, cholesterol, and overweight/obesity.