Singapore Epidemiology Of Eye Diseases Research Articles

Application of a deep-learning marker for morbidity and mortality prediction derived from retinal photographs: a cohort development and validation study

Biological ageing markers are useful to risk stratify morbidity and mortality more precisely than chronological age. In this study, we aimed to develop a novel deep-learning-based biological ageing marker (referred to as RetiPhenoAge hereafter) using retinal images and PhenoAge, a composite biomarker of phenotypic age. We used retinal photographs from the UK Biobank dataset to train a deep-learning algorithm to predict the composite score of PhenoAge. We used a deep convolutional neural network architecture with multiple layers to develop our deep-learning-based biological ageing marker, as RetiPhenoAge, with the aim of identifying patterns and features in the retina associated with variations of blood biomarkers related to renal, immune, liver functions, inflammation, and energy metabolism, and chronological age. We determined the performance of this biological ageing marker for the prediction of morbidity (cardiovascular disease and cancer events) and mortality (all-cause, cardiovascular disease, and cancer) in three independent cohorts (UK Biobank, the Singapore Epidemiology of Eye Diseases [SEED], and the Age-Related Eye Disease Study [AREDS] from the USA). We also compared the performance of RetiPhenoAge with two other known ageing biomarkers (hand grip strength and adjusted leukocyte telomere length) and one lifestyle factor (physical activity) for risk stratification of mortality and morbidity. We explored the underlying biology of RetiPhenoAge by assessing its associations with different systemic characteristics (eg, diabetes or hypertension) and blood metabolite levels. We also did a genome-wide association study to identify genetic variants associated with RetiPhenoAge, followed by expression quantitative trait loci mapping, a gene-based analysis, and a gene-set analysis. Cox proportional hazards models were used to estimate the hazard ratios (HRs) and corresponding 95% CIs for the associations between RetiPhenoAge and the different morbidity and mortality outcomes. Retinal photographs for 34 061UK Biobank participants were used to train the model, and data for 9429participants from the SEED cohort and for 3986participants from the AREDS cohort were included in the study. RetiPhenoAge was associated with all-cause mortality (HR 1·92 [95% CI 1·42-2·61]), cardiovascular disease mortality (1·97 [1·02-3·82]), cancer mortality (2·07 [1·29-3·33]), and cardiovascular disease events (1·70[1·17-2·47]), independent of PhenoAge and other possible confounders. Similar findings were found in the two independent cohorts (HR 1·67 [1·21-2·31] for cardiovascular disease mortality in SEED and 2·07 [1·10-3·92] in AREDS). RetiPhenoAge had stronger associations with mortality and morbidity than did hand grip strength, telomere length, and physical activity. We identified two genetic variants that were significantly associated with RetiPhenoAge (single nucleotide polymorphisms rs3791224 and rs8001273), and were linked to expression quantitative trait locis in various tissues, including the heart, kidneys, and the brain. Our new deep-learning-derived biological ageing marker is a robust predictor of mortality and morbidity outcomes and could be used as a novel non-invasive method to measure ageing. Singapore National Medical Research Council and Agency for Science, Technology and Research, Singapore.

Comparative Analysis of Vision Transformers and Conventional Convolutional Neural Networks in Detecting Referable Diabetic Retinopathy

ObjectiveVision Transformers (ViTs) have shown promising performance in various classification tasks previously dominated by Convolutional Neural Networks (CNNs). However, the performance of ViTs in referable Diabetic Retinopathy (DR) detection is relatively underexplored. In this study, using retinal photographs, we evaluated the comparative performances of ViTs and CNNs on detection of referable DR. DesignRetrospective study ParticipantsA total of 48,269 retinal images from the open-source Kaggle DR detection dataset, the Messidor-1 dataset and the Singapore Epidemiology of Eye Diseases (SEED) study were included. MethodUsing 41,614 retinal photographs from the Kaggle dataset, we developed five CNN (VGG19, ResNet50, InceptionV3, DenseNet201 and EfficientNetV2S) and four ViTs models (VAN_small, CrossViT_small, ViT_small and SWIN_tiny) for the detection of referable DR. We defined the presence of referable DR as eyes with moderate or worse DR. The comparative performance of all nine models was evaluated in the Kaggle internal test dataset (with 1,045 study eyes), and in two external test sets, the SEED study (5,455 study eyes) and the Messidor-1 (1,200 study eyes). Main Outcome MeasuresArea Under Receiver Operative Curve (AUC), Specificity and Sensitvity. ResultsAmong all models, the SWIN transformer displayed the highest area under the curve (AUC) of 95.7% on the internal test set, significantly outperforming the CNN models (all P<0.001). The same observation was confirmed in the external test sets, with the SWIN transformer achieving AUC of 97.3% in SEED and 96.3% in Messidor-1. When specificity level was fixed at 80% for the internal test, the SWIN transformer achieved the highest sensitivity of 94.4%, significantly better than all the CNN models (sensitivity levels ranging between 76.3% and 83.8%; all P <0.001). This trend was also consistently observed in both external test sets. ConclusionOur findings demonstrate that ViTs provide superior performance over CNNs in detecting referable DR from retinal photographs. These results point to the potential of utilizing ViT models to improve and optimize retinal photo-based deep learning for referable DR detection.

Blood biomarkers improve the prediction of prevalent and incident severe chronic kidney disease.

The prevalence of chronic kidney disease (CKD) is high. Identification of cases with CKD or at high risk of developing it is important to tailor earlyinterventions. The objective of this studywas to identify blood metabolites associated with prevalent and incident severeCKD, and to quantify the corresponding improvement in CKD detection and prediction. Data from four cohorts were analyzed: Singapore Epidemiology of Eye Diseases (SEED) (n = 8802), Copenhagen Chronic Kidney Disease(CPH) (n = 916), Singapore Diabetic Nephropathy (n = 714), and UK Biobank (UKBB) (n = 103,051). Prevalent CKD (stages 3-5) was defined as estimated glomerular filtration rate (eGFR) < 60mL/min/1.73m2; incident severe CKD as CKD-related mortality or kidney failure occurring within 10years. We used multivariable regressions to identify, among 146 blood metabolites, those associated with CKD, and quantify the corresponding increase in performance. Chronic kidney disease prevalence (stages 3-5) and severe incidence were 11.4% and 2.2% in SEED, and 2.3% and 0.2% in UKBB. Firstly, phenylalanine (Odds Ratio [OR] 1-SD increase = 1.83 [1.73, 1.93]), tyrosine (OR = 0.75 [0.71, 0.79]), docosahexaenoic acid (OR = 0.90 [0.85, 0.95]), citrate (OR = 1.41 [1.34, 1.47]) and triglycerides in medium high density lipoprotein (OR = 1.07 [1.02, 1.13]) were associated with prevalent stages 3-5 CKD. Mendelian randomization analyses suggested causal relationships. Adding these metabolites beyond traditional risk factors increased the area under the curve (AUC) by 3% and the sensitivity by 7%. Secondly, lactate (HR = 1.33 [1.08, 1.64]) and tyrosine (HR = 0.74 [0.58, 0.95]) were associated with incident severe CKD among individuals with eGFR < 90mL/min/1.73m2 at baseline. These metabolites increased the c-index by 2% and sensitivity by 5% when added to traditional risk factors. The performance improvements of CKD detection and prediction achieved by adding metabolites to traditional risk factors are modest and further research is necessary to fully understand the clinical implications of these findings.

Deep learning algorithms to detect diabetic kidney disease fromretinal photographs in multiethnic populations with diabetes.

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. 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). 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. 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.

Predictive, preventive, and personalized management of retinal fluid via computer-aided detection app for optical coherence tomography scans.

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. 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. 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. 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. The online version contains supplementary material available at 10.1007/s13167-022-00301-5.

Deep learning algorithms for automatic detection of pterygium using anterior segment photographs from slit-lamp and hand-held cameras

Background/aimsTo evaluate the performances of deep learning (DL) algorithms for detection of presence and extent pterygium, based on colour anterior segment photographs (ASPs) taken from slit-lamp and hand-held cameras.MethodsReferable pterygium...

Utilisation of poor-quality optical coherence tomography scans: adjustment algorithm from the Singapore Epidemiology of Eye Diseases (SEED) study

PurposeTo evaluate the effect of signal strength (SS) on optical coherence tomography (OCT) parameters, and devise an algorithm to adjust the effect, when acceptable SS cannot be obtained.Methods5085 individuals (9582...

Normative profiles of neuroretinal rim area in a multiethnic Asian population: the Singapore Epidemiology of Eye Diseases study

AimsTo evaluate the normative profiles for neuroretinal rim area (RA) in a multiethnic Asian population.MethodsSubjects were recruited from the Singapore Epidemiology of Eye Diseases (2009–2015) study and underwent standardised examinations....

Association of Glaucoma Risk Genes with Retinal Nerve Fiber Layer in a Multi-ethnic Asian Population: The Singapore Epidemiology of Eye Diseases Study.

PurposeGenome-wide association studies have identified several genes associated with glaucoma. However, their roles in the pathogenesis of glaucoma remain unclear, particularly their effects on retinal nerve fiber layer (RNFL) thickness. The aim of this study was to investigate the associations between the identified glaucoma risk genes and RNFL thickness.MethodsA total of 3843 participants (7,020 healthy eyes) were enrolled from the Singapore Epidemiology of Eye Diseases (SEED) study, a population-based study composing of three major ethnic groups—Malay, Indian, and Chinese—in Singapore. Ocular examinations were performed, and spectral-domain optical coherence tomography (SD-OCT) was used to measure circumpapillary RNFL thickness. We selected 35 independent glaucoma-associated genetic loci for analysis. An linear regression model was conducted to determine the association of these variants with circumpapillary RNFL, assuming an additive genetic model. We conducted association analysis in each of the three ethnic groups, followed by a meta-analysis of them.ResultsThe mean age of the included participants was 59.4 ± 8.9 years, and the mean RFNL thickesss is 92.3 ± 11.2 µm. In the meta-analyses, of the 35 glacuoma loci, we found that only SIX6 was significantly associated with reduction in global RNFL thickness (rs33912345; β = −1.116 um per risk allele, P = 1.64E-05), and the effect size was larger in the inferior RNFL quadrant (β = −2.015 µm, P = 2.9E-6), and superior RNFL quadrant (β = −1.646 µm, P = 6.54E-5). The SIX6 association were consistently observed across all three ethnic groups. Other than RNFL, we also found several genetic varaints associated with vertical cuo-to-disc ratio (ATOH7, CDKN2B-AS1, and TGFBR3-CDC7), rim area (SIX6 and CDKN2B-AS1), and disc area (SIX6, ATOH7, and TGFBR3-CDC7). The association of SIX6 rs33912345 with NRFL thickness remained similar after further adjusting for disc area and 3 other disc parameter associated SNPs (ATOH7, CDKN2B-AS1, and TGFBR3-CDC7).ConclusionsOf the 35 glaucoma identified risk loci, only SIX6 is significantly and independently associated with thinner RNFL. Our study further supports the involvement of SIX6 with RNFL thickness and pathogensis of glaucoma.

Asian-specific vertical cup-to-disc ratio cut-off for glaucoma screening: An evidence-based recommendation from a multi-ethnic Asian population.

Evidence-based guidelines are essential for glaucoma screening to work effectively. To derive a vertical cup-to-disc ratio (VCDR) cut-off for glaucoma screening in a multi-ethnic Asian population. The Singapore Epidemiology of Eye Diseases (SEED) study is a population-based study conducted from 2004 to 2011 in a single tertiary care research institute. SEED comprised of 10 033 Chinese, Malay and Indian adults aged ≥40 (response rate 75.6%). After excluding participants with a history of glaucoma medication or surgery, 9673 participants were included for analysis. A systematic eye examination, which included applanation tonometry, visual field testing, gonioscopy and dilated fundus examination was conducted. Diagnosis of glaucoma. The distribution of VCDR and VCDR asymmetry were relatively homogenous in this multi-ethnic Asian population, with a 97.5th percentile value of 0.67 and 0.17, respectively. In the absence of more definite signs of glaucoma, VCDR ≥0.60 and VCDR asymmetry ≥0.20 provided the best balance between sensitivity (95.1%) and specificity (90.9%) in detecting glaucoma. For larger optic disc (≥2.0 mm), VCDR ≥0.65 with VCDR asymmetry ≥0.20 provided the best balance between sensitivity (84.8%) and specificity (93.2%). Overall, VCDR ≥0.60 with VCDR ≥0.20 asymmetry provides a good balance between sensitivity and specificity in detecting glaucoma. For larger optic disc, VCDR ≥0.65 should be considered instead to mitigate against false-referrals due to larger physiological disc cupping. Our findings may act as a reference to populations with similar VCDR distribution.

Association Between Visual Impairment and Decline in Cognitive Function in a Multiethnic Asian Population

With the rapidly aging population, the burden of visual impairment (VI) and cognitive decline is expected to increase. Previous cross-sectional studies suggest an association between these 2 health outcomes. However, few longitudinal reports have examined this association, and to our knowledge, no studies have been performed in Asian populations. Further investigation on this association may help to better identify individuals at risk of cognitive decline. To examine the longitudinal association between VI and decline in cognitive function in a multiethnic Asian population. In this longitudinal, population-based, prospective cohort study, Chinese, Indian, and Malay adults 60 years or older at baseline were recruited from the Singapore Epidemiology of Eye Diseases (SEED) study. At baseline, participants from the SEED study were recruited under 3 studies: the Singapore Malay Eye Study (SiMES; 2004-2006), the Singapore Indian Eye Study (SINDI; 2007-2009), and the Singapore Chinese Eye Study (SCES; 2009-2011). Eligible participants were reexamined after 6 years (2011-2013 for SiMES, 2013-2015 for SINDI, and 2015-2017 for SCES). Data analysis was performed from November 1 to 24, 2019. Visual impariment was defined as presenting visual acuity worse than 20/40 based on the better-seeing eye. Cognitive function was assessed using a locally validated Abbreviated Mental Test (AMT). The association between baseline VI and change in AMT score was determined using the multivariable linear regression model adjusting for baseline age; sex; race/ethnicity; presence of diabetes, hyperlipidemia, hypertension, and chronic kidney disease; history of cardiovascular disease; smoking status; alcohol intake; body mass index; educational status; and AMT score. A total of 2478 individuals (1256 [50.7%] male; 1073 Chinese, 768 Indian, and 637 Malay adults) with mean (SD) age of 67.6 (5.6) years were evaluated, of whom 489 (19.7%) had reduction in AMT scores over 6 years. Baseline VI was associated with a decrease in AMT score over 6 years (β = -0.27; 95% CI, -0.37 to -0.17; P < .001). When change in vision over 6 years was evaluated, unchanged or deteriorated VI was associated with a decrease in AMT score over 6 years (β = -0.29; 95% CI, -0.40 to -0.18; P < .001). Among individuals with baseline VI and a substantial decrease in AMT score of 3 units or more over 6 years, the leading causes of VI were undercorrected refractive error (14 [45.2%]) and cataract (11 [35.5%]). In this study, poor vision was independently associated with a decline in cognitive function. Causes of visual loss in these cases were mostly preventable, further suggesting that preserving good vision may be an important interventional strategy for mitigating cognitive decline.

Association between Macular Thickness Profiles and Visual Function in Healthy Eyes: The Singapore Epidemiology of Eye Diseases (SEED) Study

This study aimed to evaluate the association between optical coherence tomography (OCT)-measured retinal layer thickness parameters with clinical and patient-centred visual outcomes in healthy eyes. Participants aged 40 and above were recruited from the Singapore Epidemiology of Eye Diseases Study, a multi-ethnic population-based study. Average macular, ganglion cell-inner plexiform layer (GCIPL), and outer retinal thickness parameters were obtained using the Cirrus High Definition-OCT. Measurements of best-corrected visual acuity (BCVA) and 11-item visual functioning questionnaire (VF-11) were performed. Associations between macular thickness parameters, with BCVA and Rasch-transformed VF-11 scores (in logits) were assessed using multivariable linear regression models with generalized estimating equations, adjusted for relevant confounders. 4,540 subjects (7,744 eyes) with a mean age of 58.8 ± 8.6 years were included. The mean BCVA (LogMAR) was 0.10 ± 0.11 and mean VF-11 score was 5.20 ± 1.29. In multivariable regression analysis, thicker macula (per 20 µm; β = −0.009) and GCIPL (per 20 µm; β = −0.031) were associated with better BCVA (all p ≤ 0.001), while thicker macula (per 20 µm; β = 0.04) and GCIPL (per 20 µm, β = 0.05) were significantly associated with higher VF-11 scores (all p < 0.05). In conclusion, among healthy Asian eyes, thicker macula and GCIPL were associated with better vision and self-reported visual functioning. These findings provide further understanding on the potential influence of macular thickness on visual function.

Deep Learning for Automated Sorting of Retinal Photographs

Though the domain of big data and artificial intelligence in health care continues to evolve, there is a lack of systemic methods to improve data quality and streamline the preparation process. To address this, we aimed to develop an automated sorting system (RetiSort) that accurately labels the type and laterality of retinal photographs. Cross-sectional study. RetiSort was developed with retinal photographs from the Singapore Epidemiology of Eye Diseases (SEED) study. The development of RetiSort was composed of 3 steps: 2 deep-learning (DL) algorithms and 1 rule-based classifier. For step 1, a DL algorithm was developed to locate the optic disc, the "landmark feature." For step 2, based on the location of the optic disc derived from step 1, a rule-based classifier was developed to sort retinal photographs into 3 types: macular-centered, optic disc-centered, or related to other fields. Step 2 concurrently distinguished laterality (i.e., the left or right eye) of macular-centered photographs. For step 3, an additional DL algorithm was developed to differentiate the laterality of disc-centered photographs. Via the 3 steps, RetiSort sorted and labeled retinal images into (1) right macular-centered, (2) left macular-centered, (3) right optic disc-centered, (4) left optic disc-centered, and (5) images relating to other fields. Subsequently, the accuracy of RetiSort was evaluated on 5000 randomly selected retinal images from SEED as well as on 3 publicly available image databases (DIARETDB0, HEI-MED, and Drishti-GS). The main outcome measure was the accuracy for sorting of retinal photographs. RetiSort mislabeled 48 out of 5000 retinal images from SEED, representing an overall accuracy of 99.0% (95% confidence interval [CI], 98.7-99.3). In external tests, RetiSort mislabeled 1, 0, and 2 images, respectively, from DIARETDB0, HEI-MED, and Drishti-GS, representing an accuracy of 99.2% (95% CI, 95.8-99.9), 100%, and 98.0% (95% CI, 93.1-99.8), respectively. Saliency maps consistently showed that the DL algorithm in step 3 required pixels in the central left lateral border and optic disc of optic disc-centered retinal photographs to differentiate the laterality. RetiSort is a highly accurate automated sorting system. It can aid in data preparation and has practical applications in DL research that uses retinal photographs.

Age-related changes of individual macular retinal layers among Asians

We characterized the age-related changes of the intra-retinal layers measured with spectral-domain optical coherence tomography (SD-OCT; Cirrus high-definition OCT [Carl Zeiss Meditec]. The Singapore Epidemiology of Eye Diseases is a population-based, cross-sectional study of Chinese, Malays and Indians living in Singapore. Iowa Reference Algorithms (Iowa Institute for Biomedical Imaging) were used for intra-retinal layer segmentation and mean thickness of 10 intra-retinal layers rescaled with magnification correction using axial length value. Linear regression models were performed to investigate the association of retinal layers with risk factors. After excluding participants with history of diabetes or ocular diseases, high-quality macular SD-OCT images were available for 2,047 participants (44–89 years old). Most of the retinal layers decreased with age except for foveal retinal nerve fiber layer (RNFL) and the inner/outer segments of photoreceptors where they increased with age. Men generally had thicker retinal layers than women. Chinese have the thickest RNFL and retinal pigment epithelium amongst the ethnic groups. Axial length and refractive error remained correlated with retinal layers in spite of magnification correction. Our data show pronounced age-related changes in retinal morphology. Age, gender, ethnicity and axial length need be considered when establishing OCT imaging biomarkers for ocular or systemic disease.

Compensation of retinal nerve fibre layer thickness as assessed using optical coherence tomography based on anatomical confounders

Background/AimsTo compensate the retinal nerve fibre layer (RNFL) thickness assessed by spectral-domain optical coherence tomography (SD-OCT) for anatomical confounders.MethodsThe Singapore Epidemiology of Eye Diseases is a population-based study, where 2698...

Association of Diabetes With Central Corneal Thickness Among a Multiethnic Asian Population

Thicker or thinner central corneas may lead to either overestimation or underestimation of intraocular pressure, which is the most important causal and treatable risk factor for glaucoma. However, the findings on the associations between diabetes, random glucose, and glycated hemoglobin A1c (HbA1c) with central corneal thickness (CCT) are conflicting. To evaluate the associations between diabetes, random glucose, and HbA1c with CCT in a multiethnic Asian population. Cross-sectional analysis of the Singapore Epidemiology of Eye Diseases (SEED) Study conducted from 2004 to 2011. A total of 10 033 Chinese, Malay, and Indian individuals 40 years or older residing in Singapore were recruited. Participants with incomplete information on diabetes status (448 participants), prior refractive or cataract surgery (1940 eyes), and corneal edema or dystrophy (29 eyes) were excluded. A meta-analysis was conducted to estimate the overall association of diabetes with CCT. Standardized clinical examinations and interviewer-administered questionnaire to collect information about demographic, systemic, and ocular factors. Measurement of CCT using ultrasound pachymetry. A total of 8846 adults (mean [SD] age, 57.9 [9.9] years; 4447 women [50.3%]) (17 201 eyes) were included in the final analyses. The CCT profile was similar among participants with and without diabetes (mean [SD] CCT, 545.3 [33.7] μm vs 544.8 [33.9] μm; P = .39). Following adjustments of age, sex, ethnicity, corneal curvature, axial length, and body mass index, CCT was a mean (SD) of 4.9 (0.8) μm (95% CI, 3.3-6.5 μm) thicker in patients with diabetes than those without diabetes. Multivariable analyses also showed that thicker CCT was associated with higher random glucose (per 10 mg/dL [to convert to mmol/L, multiply by 0.0555], β = 0.3; 95% CI, 0.2-0.4) and higher HbA1c (per percentage, β = 1.5; 95% CI, 1.0-2.1) (all P < .001). These associations were significant in the subgroup with diabetes but not in the subgroup without diabetes. A meta-analysis including 12 previous population- and clinical-based studies showed that CCT was 12.8 μm (95% CI, 8.2-17.5 μm) thicker in eyes of patients with diabetes. These findings suggest that diabetes and hyperglycemia were associated with thicker CCT. This study provides useful information on the interpretation of intraocular pressure in patients with diabetes.

Aldose Reductase Polymorphisms, Fasting Blood Glucose, and Age-Related Cortical Cataract.

To determine whether there is an association between polymorphisms of the AKR1B1 gene and cortical cataract in the presence of hyperglycemia. In the second cross section of the Blue Mountains Eye Study (BMES), 3508 participants (2334 at 5-year follow-up and 1174 newly recruited participants) were examined during 1997 to 2000. Cataract was graded from lens photographs using the Wisconsin Cataract Grading System. Fasting blood glucose (FBG) was measured. Continuous imputed dosages of minor alleles of 17 AKR1B1 single nucleotide polymorphisms (SNPs) were assessed for associations with prevalent cortical cataract. Gene-environment interactions between SNPs and FBG were examined. Odds ratios (OR) and 95% confidence intervals (CI) for prevalent cortical cataract were estimated using logistic regression adjusting for age, sex, smoking, hypertension, education, and myopia. A P value of 0.005 was considered statistically significant after correction for 10 independent tests. Replication of significant associations found in the BMES sample was conducted in the Singapore Epidemiology of Eye Diseases (SEED) study (n = 10,033). No polymorphism was associated with prevalent cortical cataract. A significant interaction was observed between rs9640883 and FBG (Pinteraction = 0.004), with increased cortical cataract prevalence associated with rs9640883 minor allele dosage in those with FBG >6.0 mM (strata-specific OR 1.72, 95% CI 1.09-2.72). No similar association was found in participants with normal FBG (OR 0.85, 95% CI 0.69-1.04). This interaction was not evident in the SEED study. The identified interaction between rs9640883 and FBG in relation to cortical cataract was not replicated but may warrant further investigation.

Prevalence, Risk Factors, and Impact of Myopic Macular Degeneration on Visual Impairment and Functioning Among Adults in Singapore.

To determine the prevalence, risk factors, and impact of myopic macular degeneration (MMD) on visual impairment and functioning among adults in Singapore. A comprehensive eye examination, including subjective refraction, axial length, and visual acuity (VA) measurements, was performed in adults aged ≥40 years in the Singapore Epidemiology of Eye Diseases (SEED) study. From fundus photographs, MMD was graded using the International META-PM classification. Vision-specific functioning (VSF) was assessed with a validated visual-functioning questionnaire (VF-11) using Rasch analysis. A total of 8716 phakic subjects were included in this analysis. The mean age (± SD) was 57.2 ± 9.5 years (33.5% Malays, 33.2% Indians, and 33.3% Chinese). The prevalence of myopia (spherical equivalent [SE] ≤ -0.5 diopters [D]) and high myopia (SE ≤ -5.0 D) was 35.7% and 6.0%, respectively. The age-standardized prevalence of MMD was 3.8% (95% confidence interval [CI], 3.4-4.3%). The prevalence of MMD was 7.7% among low to moderate myopes, and 28.7% among high myopes. The prevalence of MMD increased nonlinearly with SE and age. MMD was associated with older age, more myopic SE, and lower education. Subjects with Meta-PM categories 3 or 4 in the better-seeing eye had worse best-corrected VA (β, 0.19; 95%CI, 0.16-0.23) and poorer VSF (β, -9.7; 95%CI, -17.6 to -1.8) than those without MMD after multivariate adjustments. Approximately 1 in 26 phakic adults in Singapore has MMD. Older age and myopic SE are major risk factors of MMD. Severe MMD has a substantial impact on visual impairment and functioning.

Direct and Indirect Associations Between Diabetes and Intraocular Pressure: The Singapore Epidemiology of Eye Diseases Study.

The association between diabetes and IOP is controversial; diabetes is associated with thicker central corneal thickness (CCT), and thicker CCT is associated with higher IOP. We therefore aimed to clarify the diabetes-IOP association, considering CCT as a potential mediator. We included 8636 participants from the Singapore Epidemiology of Eye Diseases (SEED) Study. Associations of diabetes, serum glucose, or HbA1c with IOP were assessed using regressions models, with adjustments for potential confounding factors. Regression-based mediation (path) analyses were further performed to evaluate the indirect effects of diabetes on IOP through the mediator (CCT), in addition to the direct effect of diabetes on IOP. Of the 8636 participants, 2524 (29.23%) had diabetes. Diabetes, higher serum glucose, or HbA1c levels were all associated with higher IOP (all P < 0.01). The effect of diabetes on IOP was partially and minimally mediated through CCT; the proportion of mediating effect of CCT was 11.09% of the total effect of diabetes on IOP. Axial length and spherical equivalent were not mediating variables in the diabetes-IOP association. Findings were consistent across three ethnicity groups. Diabetes or higher long-term hyperglycemia was associated with higher IOP. CCT contributed a small proportion of mediating effect to the total effect of diabetes on IOP. We conclude that high IOP observed in diabetes is mainly due to the direct association of diabetes and IOP, and this finding may have pathophysiologic significance with respect to the risk of glaucoma among persons with diabetes.

Projection of Eye Disease Burden in Singapore

Singapore's ageing population is likely to see an increase in chronic eye conditions in the future. This study aimed to estimate the burden of eye diseases among resident Singaporeans stratified for age and ethnicity by 2040. Prevalence data on myopia, epiretinal membrane (ERM), retinal vein occlusion (RVO), age macular degeneration (AMD), diabetic retinopathy (DR), cataract, glaucoma and refractive error (RE) by age cohorts and educational attainment from the Singapore Epidemiology of Eye Diseases (SEED) study were applied to population estimates from the Singapore population model. All eye conditions are projected to increase by 2040. Myopia and RE will remain the most prevalent condition, at 2.393 million (2.32 to 2.41 million) cases, representing a 58% increase from 2015. It is followed by cataract and ERM, with 1.33 million (1.31 to 1.35 million), representing an 81% increase, and 0.54 million (0.53 to 0.549 million) cases representing a 97% increase, respectively. Eye conditions that will see the greatest increase from 2015 to 2040 in the Chinese are: DR (112%), glaucoma (100%) and ERM (91.4%). For Malays, DR (154%), ERM (136%), and cataract (122%) cases are expected to increase the most while for Indians, ERM (112%), AMD (101%), and cataract (87%) are estimated to increase the most in the same period. Results indicate that the burden for all eye diseases is expected to increase significantly into the future, but at different rates. These projections can facilitate the planning efforts of both policymakers and healthcare providers in the development and provision of infrastructure and resources to adequately meet the eye care needs of the population. By stratifying for age and ethnicity, high risk groups may be identified and targeted interventions may be implemented.