Regional Disparities in Prevalence and Causes of Visual Impairment and Blindness, Cataract Surgical Coverage and Ocular Findings in Adults 50 Years and Older in Brazil: A Comparative Analysis between Two Population-Based Cross-Sectional Surveys Performed Within a 10-Year Interval.

PURPOSE: To investigate disparities in the prevalence and causes of visual impairment and blindness, cataract surgical coverage, and ocular findings in older adults from two Brazilian geo-socio-demographic areas, São Paulo and Parintins cities. METHODS: Data from two population-based studies including participants 50 years and older from the cities of São Paulo (São Paulo Eye Study - SPES, 2004) and Parintins (Brazilian Amazon Region Eye Survey - BARES, 2014) were aggregated. RESULTS: A total of 5318 participants (3677 from SPES;1641 from BARES) were included. The prevalence of severe visual impairment (SVI) and blindness were, respectively, 0.74% (0.46-1.02) and 0.77% (0.48-1.05) in SPES and 1.72% (1.09-2.35) and 3.44% (2.55-4.33) in BARES. SVI and blindness were associated with BARES study [OR = 2.27 (1.30-3.95); p = .004 - SVI] [OR:4.07 (2.51-6.60); p < .001- blindness]; and older age [OR = 10.93 (4.20-28.45); p < .001 - SPES; OR = 17.96 (8.75-36.83); p < .001 - BARES] while higher education level was a protective factor [OR = 0.21 (0.05-0.95) - SPES; p = .042; OR = 0.21 (0.05-0.91); p = .037 - BARES]. Cataract was the main cause of bilateral severe visual impairment (25.93% in SPES and 64.29% in BARES) and bilateral blindness (21.43% in SPES and 35.71% in BARES). Cataract surgical coverage was significantly lower in BARES (36.32%) compared to SPES (57.75%). CONCLUSION: The prevalence of SVI and blindness was three times higher in older adults from the Brazilian Amazon compared to those living in São Paulo city, despite a 10-year interval between the two studies. These disparities should be mitigated by initiatives to promote access to eye care services targeting underprivileged and remote Brazilian areas.

Impact of pterygium on central corneal thickness measured by optical coherence tomography in older adults.

PURPOSE: To measure the central corneal thickness (CCT) using anterior segment optical coherence tomography (AS-OCT) in older adults with and without pterygium from the Brazilian Amazon Region Eye Survey (BARES). METHODS: BARES is a population-based epidemiological cross-sectional study conducted in Parintins city. Participants were residents ≥45 years of age identified through a door-to-door interview. Eligible participants were invited for a comprehensive eye exam. Pterygium occurrence and severity were assessed by ophthalmologists through slit-lamp examination considering its location (nasal or/and temporal) and severity (lesion with extension <3 mm, ≥3 mm not reaching the pupillary margin or ≥3 mm reaching the pupillary margin). CCTs were obtained and measurements from the more severely affected eye were included. Images were analyzed offline by masked observers. RESULTS: A total of 671 subjects, 533 (79.4%) with pterygium in at least one eye and 138 (20.6%) without pterygium in either eye, were examined. The mean CCT evaluated by multiple linear regression and adjusted for demographic variables and pterygium severity was 521 ± 34 µm (median = 521; range = 304-665). Decreased CCT was significantly associated with age and pterygium severity. Individuals aged 65-74 years had CCT 7 µm thinner than those aged 45-54 years (p = 0.044), individuals aged 75 years and older had CCT 15 µm thinner than those aged 45-54 years (p = 0.001), and eyes with severe pterygium had CCT 33 µm thinner than eyes without pterygium (p < 0.001). CONCLUSIONS: The CCT analysis in this population-based sample shows that a thinner cornea is associated with pterygium severity and older age.

Diabetic retinopathy classification for supervised machine learning algorithms.

BACKGROUND: Artificial intelligence and automated technology were first reported more than 70 years ago and nowadays provide unprecedented diagnostic accuracy, screening capacity, risk stratification, and workflow optimization. Diabetic retinopathy is an important cause of preventable blindness worldwide, and artificial intelligence technology provides precocious diagnosis, monitoring, and guide treatment. High-quality exams are fundamental in supervised artificial intelligence algorithms, but the lack of ground truth standards in retinal exams datasets is a problem. MAIN BODY: In this article, ETDRS, NHS, ICDR, SDGS diabetic retinopathy grading, and manual annotation are described and compared in publicly available datasets. The various DR labeling systems generate a fundamental problem for AI datasets. Possible solutions are standardization of DR classification and direct retinal-finding identifications. CONCLUSION: Reliable labeling methods also need to be considered in datasets with more trustworthy labeling.

Overweight/obesity in adolescents with type 1 diabetes belonging to an admixed population. A Brazilian multicenter study.

BACKGROUND: To determine the prevalence of overweight/obesity and associated risk factors in Brazilian adolescents with type 1 diabetes (T1D) and its association with diabetic retinopathy (DR) and chronic kidney disease (CKD). METHODS: This study was performed in 14 Brazilian public clinics in ten cities, with 1,760 patients. 367 were adolescents (20.9%):184 females (50.1%), 176 (48.0%) Caucasians, aged 16.4 ± 1.9 years, age at diagnosis 8.9 ± 4.3 years, diabetes duration 8.1 ± 4.3 years, school attendance 10.9 ± 2.5 years and HbA1c 9.6 ± 2.4%. RESULTS: 95 (25.9%) patients presented overweight/obesity, mostly females. These patients were older, had longer diabetes duration, higher levels of total and LDL-cholesterol, higher prevalence of family history of hypertension, hypertension, undesirable levels of LDL-cholesterol, and metabolic syndrome compared to eutrophic patients. No difference was found regarding ethnicity, HbA1c, uric acid, laboratorial markers of non-alcoholic fatty liver disease (alanine aminotransferase, aspartate aminotransferase, gamma-glutamyl transferase). CONCLUSIONS: Almost one quarter of our patients presented overweight/obesity. These patients had higher prevalence of traditional risk factors for micro and macrovascular diabetes-related chronic complications such as diabetes duration, hypertension, high levels of LDL-cholesterol and metabolic syndrome. The majority of the patients with or without overweight/obesity presented inadequate glycemic control which is also an important risk factor for micro and macrovascular diabetes-related chronic complications. No association was found between overweight/obesity with diabetic CKD, DR and laboratorial markers of non-alcoholic fatty liver disease. The above-mentioned data point out that further prospective studies are urgently needed to establish the clinical prognosis of these young patients.

Diabetic Retinopathy Screening Using Artificial Intelligence and Handheld Smartphone-Based Retinal Camera.

BACKGROUND: Portable retinal cameras and deep learning (DL) algorithms are novel tools adopted by diabetic retinopathy (DR) screening programs. Our objective is to evaluate the diagnostic accuracy of a DL algorithm and the performance of portable handheld retinal cameras in the detection of DR in a large and heterogenous type 2 diabetes population in a real-world, high burden setting. METHOD: Participants underwent fundus photographs of both eyes with a portable retinal camera (Phelcom Eyer). Classification of DR was performed by human reading and a DL algorithm (PhelcomNet), consisting of a convolutional neural network trained on a dataset of fundus images captured exclusively with the portable device; both methods were compared. We calculated the area under the curve (AUC), sensitivity, and specificity for more than mild DR. RESULTS: A total of 824 individuals with type 2 diabetes were enrolled at Itabuna Diabetes Campaign, a subset of 679 (82.4%) of whom could be fully assessed. The algorithm sensitivity/specificity was 97.8 % (95% CI 96.7-98.9)/61.4 % (95% CI 57.7-65.1); AUC was 0·89. All false negative cases were classified as moderate non-proliferative diabetic retinopathy (NPDR) by human grading. CONCLUSIONS: The DL algorithm reached a good diagnostic accuracy for more than mild DR in a real-world, high burden setting. The performance of the handheld portable retinal camera was adequate, with over 80% of individuals presenting with images of sufficient quality. Portable devices and artificial intelligence tools may increase coverage of DR screening programs.

Diabetic Macular Edema Screened by Handheld Smartphone-based Retinal Camera and Artificial Intelligence.

Our aim was to assess the tomographic presence of diabetic macular edema in type 2 diabetes patients screened for diabetic retinopathy with color fundus photographs and an artificial intelligence algorithm. Color fundus photographs obtained with a low-cost smartphone-based handheld retinal camera were analyzed by the algorithm; patients with suspected macular lesions underwent ocular coherence tomography. A total of 366 patients were screened; diabetic macular edema was suspected in 34 and confirmed in 29 individuals, with average age 60.5 ± 10.9 years and glycated hemoglobin 9.8 ± 2.4%; use of insulin, statins, and aspirin were reported in 44.8%, 37.9%, and 34.5% of individuals, respectively; systemic blood hypertension, dyslipidemia, abdominal obesity, chronic kidney disease, and risk for diabetic foot ulcers were present in 100%, 58.6%, 62.1%, 48.3%, and 27.5% of individuals, respectively. Proliferative diabetic retinopathy was present in 31% of patients with macular edema; severity level was associated with albuminuria (p = 0.028). Eyes with macular edema had average central macular thickness 329.89 ± 80.98 m[Formula: see text]; intraretinal cysts, sub retinal fluid, hyper-reflective foci, epiretinal membrane, and vitreomacular traction were found in 87.2%, 6.4%, 85.1%, 10.6%, and 6.4% of eyes, respectively. Diabetic retinopathy screening overwhelms health systems and is typically based on color fundus photographs, with high false-positive rates for the detection of diabetic macular edema. The present, semi-automated strategy comprising artificial intelligence algorithms integrated with smartphone-based retinal cameras could improve screening in low-resource settings with limited availability of ocular coherence tomography, allowing increased access rates and ultimately contributing to tackle preventable blindness.

Human Leukocyte Antigens class II (HLA II) gene profile from an admixed population of patients with type 1 diabetes with severe diabetic retinopathy: a nested case-control study in Brazil.

BACKGROUND: Although the well-established role of the HLA genes on the predisposition of type 1 diabetes (T1D), its contribution to the development and progression of diabetic retinopathy is still unclear, especially in admixed populations. We aimed to study the relationship between HLA alleles and severe diabetic retinopathy in a highly admixed population of T1D patients. METHODS: This was a nested case-control study based on a cross-sectional, nationwide survey conducted in Brazil. We included 117 patients with severe diabetic retinopathy and 117 random controls composed of T1D patients without retinopathy, matched for diabetes duration. HLA-class II genes (HLA-DRB1, -DQA1, and -DQB1) were genotyped using the SSO and NGS methods. RESULTS: Haplotypes HLA-DRB1*04:05 ~ DQA1*03:01 g ~ DQB1*03:02 (OR 1.75, CI 0.97-3.16, p value 0.058) and HLA-DRB1*13:02 ~ DQA1*01:02 ~ DQB1*06:04 (OR 5.18, CI 1.12-23.09, p value 0.019) were more prevalent on the severe DR group but they did not present statistically difference after Bonferroni correction. The most frequent haplotype on both groups was HLA-DRB1*03:01 ~ DQA1*05:01 g ~ DQB1*02:01 (29.6% on severe DR and 33.33% on the control group). CONCLUSIONS: Our study showed no influence of HLA genes on the development of DR. Further longitudinal data is needed to better understand the role of genetic factors on this multifactorial significant microvascular complication.

Diabetes-related chronic complications in Brazilian adolescents with type 1 diabetes. A multicenter cross-sectional study.

AIMS: To investigate the prevalence of diabetes-related chronic complications (DRCCs) and its associated factors in Brazilian adolescents with type 1 diabetes (T1D). METHODS: This nationwide study was conducted in 14 public clinics in 10 cities, with 1,760 patients, 367 adolescents, with 328 eligible for this study. Evaluated DRCCs were retinopathy (DR), chronic kidney disease (CKD), peripheral neuropathy (DPN) and cardiovascular autonomic neuropathy (CAN). RESULTS: Among eligible patients, 184 were females (50.1%), age range 13-19 years, HbA1c 9.6% ± 2.4, aged 8.9 ± 4.3 years at diagnosis and diabetes duration of 8.1 ± 4.3 years. 103 (31.4%) patients presented any type of DRCC. CKD was found in 46 (14.0%), CAN in 41(12.5%), DR in 28 (8.5%) and DPN in 16 (4.9%) patients. One, two or three DRCCs were observed in 79 (24.1%), 19 (5.8%) and 5 (1.5%) patients, respectively, and were associated with longer diabetes duration, higher HbA1c and diastolic blood pressure levels (dBP), use of renin angiotensin inhibitors and lower adherence to diet. CONCLUSIONS: A high percentage of patients presented some kind of DRCC, associated with diabetes duration, glycemic control, dBP, adherence to diet. Educational programs should start from the diagnosis to avoid DRCCs in this young population.

Relationship between Proliferative Diabetic Retinopathy and Inflammatory Markers in Patients with Type 1 Diabetes in Brazil: A Nested Case Control Study.

INTRODUCTION: This study examined the relationship between proliferative diabetic retinopathy (PDR) and serum levels of C-reactive protein, VEGF, TNF-α, and IL-6 inflammatory biomarkers, related to the pathophysiology of diabetic retinopathy. METHODS: This cross-sectional, case control study comprised 240 patients with type 1 diabetes (80 cases with PDR and 160 controls without diabetic retinopathy) who were matched for gender and duration of diabetes. RESULTS: C-reactive protein was the only inflammatory biomarker that was positively related to PDR (OR 1.96; 95% CI 1.01-3.78, p = 0.0045). We also noted an association between high glycated hemoglobin levels, the use of angiotensin-converting enzyme inhibitor, low glomerular filtration rate, and PDR. CONCLUSION: Patients with higher levels of C-reactive protein are more likely to present with PDR. We did not find a link between serum levels of VEGF, TNF-α, or IL-6 and PDR. The function of inflammatory biomarkers in PDR must be addressed in further studies.

Genomic ancestry as a risk factor for diabetic retinopathy in patients with type 1 diabetes from an admixed population: a nested case-control study in Brazil.

AIMS: The influence of genetic factors on the development and progression of diabetic retinopathy is still unclear. Previous studies showed controversial results. We aimed to characterize the relationship between genomic ancestry and self-reported color/race with severe diabetic retinopathy in patients with type 1 diabetes belonging to a highly admixed population. METHODS: This study was a nested case-control based on data collected from a large cross-sectional, nationwide survey conducted in clinics from all five geographic regions of Brazil. For the present study, we included 414 individuals. Cases (n = 176) were considered if they had severe non-proliferative or proliferative diabetic retinopathy, and controls (n = 238) were type 1 diabetes patients without retinopathy, matched for diabetes duration by a range of 5 years. Indirect ophthalmoscopy was performed, and individual genomic ancestry was inferred using a panel of 46 ancestry informative markers. RESULTS: The backward stepwise logistic regression analysis showed that African genomic ancestry (OR 3.9, p = 0.045), HbA1c (OR 1.24, p = 0.001), glomerular filtration rate (OR 0.98, p < 0.001) and hypertension (OR 2.52, p < 0.001) were associated with severe diabetic retinopathy after adjusting for clinical and demographic data. Self-reported color/race was not statistically associated with diabetic retinopathy. CONCLUSIONS: Genomic ancestry, as well as clinical variables such as hypertension, impaired glomerular filtration rate and poor diabetes control (HbA1c), was important risk factor for the development of severe diabetic retinopathy. Further studies are needed, especially in highly admixed populations, to better understand the role of genomic ancestry and possible genes that might be associated with the development and/or progression of diabetic retinopathy.

Diabetic Retinopathy May Indicate an Increased Risk of Cardiovascular Disease in Patients With Type 1 Diabetes-A Nested Case-Control Study in Brazil.

Objective: Cardiovascular disease, the leading cause of death worldwide, and diabetic retinopathy, the main cause of blindness in economically active populations, share clinical risk factors, and pathophysiological features. The aim of this study is to examine the association between diabetic retinopathy, cardiovascular disease, and common risk factors in patients with type 1 diabetes. Design and Methods: This nested case-control study was performed in patients from the Brazilian Type 1 Diabetes Study Group, a nationwide survey that was conducted in Brazil and enrolled 1,760 patients with type 1 diabetes. A total of 342 patients were selected (57 cases with macrovascular disease and 285 controls who were matched for duration of diabetes and gender). Results: In the exploratory analysis, stratified by cardiovascular disease, the following variables were statistically significant: age (p=0.037), hypertension (p=0.035), high BMI (p = 0.046), diabetic retinopathy (p = 0.003), and chronic kidney disease (p = 0.026). By multivariate logistic regression, patients with diabetic retinopathy were more likely to develop cardiovascular disease (OR 2.16, 95% CI 1.16-4.02, p = 0.015). Although to a lesser extent than diabetic retinopathy, higher BMI levels were also related to an increase in the risk of cardiovascular disease of 1.08 (95% CI 1.01-1.15, p = 0.024). Conclusion: The presence of diabetic retinopathy indicates a greater risk for cardiovascular disease in Brazilian patients with type 1 diabetes. Further studies are warranted to determine whether a noninvasive exam, such as fundoscopy, could help identify patients who show an increased risk for cardiovascular disease.

Population-Based Cataract Surgery Complications and Their Impact on Visual Status in the Brazilian Amazon Region.

PURPOSE: To investigate cataract surgery prevalence, complications, visual outcomes, and coverage in a population of older adults from the Brazilian Amazon region. DESIGN: Population-based cross-sectional study. METHODS: Individuals ≥45 years of age from urban and rural areas of Parintins City, Brazil, were identified by cluster random sampling. Participants underwent ophthalmic examination, including uncorrected visual acuity, acuity with presenting correction (APC), and best-corrected visual acuity (BCVA) from each eye. In those with previous cataract surgery, surgical technique and complications including its contribution to vision impairment were noted. The association of surgical complications and surgical coverage with sociodemographic variables was assessed by multiple logistic regression. RESULTS: A total of 2,384 persons were determined to be eligible, and 2,041 (85.6%) were examined. Overall, prevalence of cataract surgery was 8.6%, with 11.3% urban and 5.0% rural. Surgical complications were associated with males, older age, and urban residency and were found in 60.4%, with posterior capsule opacification in 29.6% and posterior capsule rupture in 16.3%. Among the 270 eyes having cataract surgery, APC ≥20/32 was found in 44.4%, APC 20/40 to 20/63 in 20.8%, APC <20/63 to 20/200 in 14.4%, APC <20/200 to 20/400 in 2.6%, and APC <20/400 in 17.8%. Low surgical coverage was found with 57.9% of those with visual impairment due to cataract not being treated surgically. CONCLUSIONS: Although a reasonable prevalence of cataract surgery was found, a high complication rate, poor visual outcomes, and low cataract surgery coverage indicated that actions to improve quality and appropriate postsurgical management should be planned and implemented by public health authorities.

Current epidemiology of diabetic retinopathy in patients with type 1 diabetes: a national multicenter study in Brazil.

BACKGROUND: Diabetic retinopathy is the leading cause of blindness in economically active populations. The aims of this study were to estimate the prevalence and to identify risk factors for diabetic retinopathy in patients with type 1 diabetes in Brazil. METHODS: This was a nationwide, cross-sectional study conducted between August 2010 and August 2014. The study included 1760 patients with type 1 diabetes. Patients underwent a standard questionnaire, clinical and laboratory analyses and were screened for diabetic retinopathy. To analyze the risk factors related to diabetic retinopathy, two models of logistic regression models were performed, one considering vision-threatening cases and the other with any diabetic retinopathy cases as dependent variables. The group with vision-threatening included patients with severe non-proliferative diabetic retinopathy, proliferative diabetic retinopathy and macular edema. RESULTS: In total, 1644 patients (mean age, 30.1± 12.0 years; duration of diabetes, 15.3 ± 9.3 years; female, 55.8%) were studied. 35.7% presented diabetic retinopathy and 12% presented vision-threatening diabetic retinopathy. Three risk factors associated with diabetic retinopathy were in common to both groups: longer diabetes duration (OR 1.07; 95% CI, 1.05-1.09), higher levels of HbA1c (OR 1.24; CI, 1.17-1.32) and higher levels of serum uric acid (OR 1.22; CI, 1.13-1.31) (p < 0.001 for all comparisons). CONCLUSION: The higher rate of vision-threatening retinopathy found in our study highlights the need to improve access to eye care and screening programs for diabetic retinopathy in Brazil. In addition to traditional risk factors, we found an association between serum uric acid levels and diabetic retinopathy. Further studies are needed to address this association.

Presbyopia and Ocular Conditions Causing Near Vision Impairment in Older Adults From the Brazilian Amazon Region.

PURPOSE: To investigate the prevalence and causes of near vision impairment (NVI) in a population of older adults from the Brazilian Amazon Region. DESIGN: Population-based cross-sectional study. METHODS: Cluster sampling was used in randomly selecting subjects 45 years of age and older from urban and rural areas of Parintins city, Brazil. Participants underwent ophthalmic examination, including uncorrected (UCNVA), presenting (PNVA), and best-corrected near visual acuity (BCNVA) from each eye; biomicroscopy; funduscopy; and subjective refraction, including testing with additional lenses for near vision optical correction. A principal cause for NVI was assigned by the ophthalmologist and presbyopia was defined as UCNVA ≤ 20/40 changing to > 20/40 with BCNVA. Free-of-charge glasses were provided for those in need. RESULTS: A total of 2384 subjects were enumerated and 2025 had reliable NVA measurements from both eyes. The prevalence of NVI in the better-seeing eye was 96.5% with UCNVA, decreasing to 81.1% with PNVA and to 20.5% with BCNVA. Presbyopia was the principal cause of NVI in 71.8%, followed by cataract (16.5%) and pterygium (2.5%), and was associated with younger age and high schooling. Glasses for near vision were prescribed and provided to 1414 (69.8%) participants. CONCLUSIONS: A high prevalence of NVI was detected even in those wearing glasses for near. Prescription and provision of low-cost reading glasses should be considered by Brazilian health authorities to address this easily and promptly correctable form of vision impairment.

Regional differences in the prevalence of diabetic retinopathy: a multi center study in Brazil.

BACKGROUND: Diabetic retinopathy has a significant impact in every healthcare system. Despite that fact, there are few accurate estimates in the prevalence of DR in Brazil's different geographic regions, particularly proliferative DR and diabetic macular edema. This study aims to determine the prevalence of diabetic retinopathy in Brazil's five continental regions and its determinant factors. METHODS: This multi center, cross-sectional, observational study, conducted between August 2011 and December 2014, included patients with type 1 diabetes from the 5 Brazilian geographic regions (South, Southeast, North, Northeast and Midwest). During a clinical visit, a structured questionnaire was applied, blood sampling was collected and each patient underwent mydriatic binocular indirect ophthalmoscopy evaluation. RESULTS: Data was obtained from 1644 patients, aged 30.2 ± 12 years (56.1% female, 54.4% Caucasian), with a diabetes duration of 15.5 ± 9.3 years. The prevalence of diabetic retinopathy was 242 (36.1%) in the Southeast, 102 (42.9%) in the South, 183 (29.9%) in the North and Northeast and 54 (41.7%) in the Midwest. Multinomial regression showed no difference in the prevalence of non-proliferative diabetic retinopathy in each geographic region, although, prevalence of proliferative diabetic retinopathy (p = 0.022), and diabetic macular edema (p = 0.003) was higher in the Midwest. Stepwise analyses reviled duration of diabetes, level of HbA1c and hypertension as independent variables. CONCLUSIONS: The prevalence of non proliferative diabetic retinopathy in patients with type 1 diabetes was no different between each geographic region of Brazil. The Midwest presented higher prevalence of proliferative diabetic retinopathy and diabetic macular edema. Duration of DM and glycemic control is of central importance to all. Hypertension is another fundamental factor to every region, at special in the South and Southeast. Glycemic control and patients in social and economic vulnerability deserves special attention in the North and Northeast of Brazil.