Inter-grader reliability in the Danish screening programme for diabetic retinopathy

PURPOSE

The Danish Registry of Diabetic Retinopathy includes information from >200 000 patients who attends diabetic retinopathy (DR) screening in Denmark. Screening of patients with uncomplicated type 2 diabetes is often performed by practicing ophthalmologists, while patients with type 1 and complicated type 2 diabetes attends screening at hospitals. We performed a clinical reliability study of retinal images from Danish screening facilities to explore the inter-grader agreement between the primary screening ophthalmologist and a blinded, certified grader.

METHODS

Invitations to participate were sent to screening facilities across Denmark. The primary grader uploaded fundus photographs with information on estimated level of DR (International Clinical Diabetic Retinopathy scale as 0 [no DR], 1-3 [mild, moderate or severe nonproliferative DR {NPDR}], or 4 [proliferative DR {PDR}]), region of screening, image style, and screening facility. Images were then regraded by a blinded, certified, secondary grader. Weighted kappa analysis was performed to evaluate agreement.

RESULTS

Fundus photographs from 230 patients (458 eyes) were received from practicing ophthalmologists (52.6%) and hospital-based grading centres (47.4%) from all Danish regions. Reported levels of DR by the primary graders were 66.8%, 12.2%, 13.1%, 1.3% and 5.5% for DR levels 0-4. The overall agreement between primary and secondary graders was 93% (κ = 0.83). Based on screening facility agreement was 96% (κ = 0.89) and 90% (κ = 0.76) for practicing ophthalmologists and hospital-based graders.

CONCLUSION

In this nationwide study, we observed a high overall inter-grader agreement and based on this, it is reasonable to assume that reported DR gradings in the screening programme in Denmark, accurately reflect the truth.

Presence and development of diabetic retinopathy in 153 238 patients with type 2 diabetes in the Danish Registry of Diabetic Retinopathy

PURPOSE

The purpose of the study was to evaluate the prevalence and incidence of diabetic retinopathy (DR) along with associated markers in patients with type 2 diabetes in the Danish DR-screening programme.

METHODS

We included all persons with type 2 diabetes in the Danish Registry of Diabetic Retinopathy, who had attended at least one episode of DR screening in 2013-2018. DR was classified as levels 0-4 indicating increasing severity. Data were linked with various national health registries to retrieve information on diabetes duration, marital status, comorbidity and systemic medication.

RESULTS

Among 153 238 persons with type 2 diabetes, median age and duration of diabetes were 66.9 and 5.3 years and 56.4% were males. Prevalence and 5-year incidences of DR, 2-step-or-more progression of DR and progression to proliferative DR (PDR) were 8.8%, 3.8%, 0.7% and 0.2%, respectively. In multivariable models, leading markers of incident DR and progression to PDR were duration of diabetes (HR 1.98, 95% CI 1.87-2.09; HR 2.89, 95% CI 2.34-3.58 per 10 years of duration) and use of insulin (HR 1.88, 95% CI 1.76-2.01; HR 2.40, 95% CI 1.84-3.13), while the use of cholesterol-lowering medicine was a protecting marker (HR 0.87, 95% CI 0.81-0.93; HR 0.70, 95% CI 0.52-0.93). From 2013 to 2015, 3-year incidence rates of PDR decreased from 1.22 to 0.45 events per 1000 person-years.

CONCLUSION

Nationally, among Danish individuals with type 2 diabetes attending DR screening, we identified duration of diabetes and use of insulin as the most important predictor for the development of DR, while cholesterol-lowering medicine was a protective factor.

Bidirectional 5-year risks of diabetic retinopathy, glaucoma and/or ocular hypertension: Results from a national screening programme

PURPOSE

We aimed to investigate if diabetic retinopathy (DR), glaucoma and/or ocular hypertension (OHT) are prospectively linked, as previous studies have proposed cross-sectional associations, but longitudinal data from larger cohorts are lacking.

METHODS

We performed a bidirectional 5 years prospective, registry-based cohort study. We extracted data from national registers, including the Danish Registry of Diabetic Retinopathy, the Danish Civil Registration System, the Danish National Patient Register and the Danish National Prescription Registry. DR level was defined by the highest level of the two eyes. Glaucoma and/or OHT was defined by diagnostic codes (H40*) or at least three redeemed prescriptions of glaucoma medication (S01E*) within 1 year. We included 205 970 persons with diabetes and 1 003 170 age- and gender-matched non-diabetes controls. Exposures were level-specific DR (i) and glaucoma and/or OHT (ii), and outcomes were hazard ratios (HRs) for 5 years incident glaucoma and/or OHT (i) and DR (ii).

RESULTS

Persons with diabetes were more likely to develop glaucoma and/or OHT (multivariable adjusted HR 1.11, 95% CI 1.06-1.15), but this did not depend on the level of DR. In persons with diabetes, those with glaucoma and/or OHT were more likely to develop DR (multivariable adjusted HR 1.12, 95% CI 1.03-1.23) within 5 years.

CONCLUSION

In a national cohort, diabetes associated with a little higher risk of upcoming glaucoma and/or OHT, and, inversely, the presence of the latter predicted a higher risk of incident DR. Nevertheless, our data do not seem to justify including glaucoma evaluation in the national Danish DR-screening programme.

Choroidal thickness and myopia in relation to physical activity - the CHAMPS Eye Study

PURPOSE

To describe the relationship between choroidal thickness (CT) and myopia in relation to physical activity (PA) in a population-based child cohort.

METHODS

In a prospective study of 307 children from the CHAMPS Study Denmark, we used objective data from GT3X accelerometer worn at four periods between 2009 and 2015 to determine the amount and intensity of PA. Intensity was estimated as counts/minutes, and cut-off points were defined at four intensity levels. Eye examinations were performed in 2015 and included autorefraction in cycloplegia, axial length (AL) by biometric and fovea-centred enhanced depth imaging optical coherence tomography. By a semi-automated method, we measured the CT at 17 targets per eye representing anatomically different locations (subfoveal, 1 and 3 millimetre in each direction of fovea).

RESULTS

Mean age at the eye examination was 15.4 ± 0.7 years. The mean AL was 23.5 ± 0.9 mm, and the mean subfoveal CT was 369 ± 87 μm. Choroidal thickness (CT) was 331 ± 68 μm for the overall macula, 355 ± 78 μm for the 1-mm zone and 304 ± 60 μm for the 3-mm zone. All CT measurements were thinner in myopic eyes (p < 0.0001) and in boys (p < 0.05). We found no association between total PA and the CT by either mixed model analysis (p = 0.074) or linear regression by any intensity levels (p = 0.22, p = 0.15 and p = 0.43).

CONCLUSION

Among adolescents aged 14-17 years, there was no association between objective PA exposures and the CT, AL or refractive error.

Retinal vascular diameters in relation to physical activity in Danish children - The CHAMPS Eye Study

Our objective was to determine associations between retinal vascular caliber and physical activity (PA) in a school-based child cohort. In a prospective study, we created a childhood cumulative average PA-index using objectively measured PA (accelerometry) assessed at four periods between 2009 and 2015. Cumulative exposure to PA intensities was estimated. Cross-sectional examinations on biomarkers, anthropometry, and ophthalmological data including retinal fundus photographs were performed in 2015. Semi-automated measurements of retinal vascular diameters were performed and summarized into central retinal arteriolar and venular equivalents (CRAE, CRVE). We included 307 participants. Mean age in 2015 was 15.4 years (0.7). The mean CRAE and CRVE were 156.5 μm (2.8) and 217.6 μm (7.7), respectively. After adjusting for age, gender, and axial length, more time in PA was independently related to thinner retinal venules (β-coefficient = -1.25 μm/%, 95% confidence interval = -2.20, -0.30, P < .01). Sedentary time was associated with wider venules (P < .01). Furthermore, birthweight (β-coefficient = 0.56 μm/%, 95% confidence interval = 0.18, 0.95, P < .01) was associated with CRVE. Blood pressure was associated with thinner retinal arterioles (β-coefficient = -0.19 μm/mmHg, 95% confidence interval = -0.36, -0.01, P = .04). We concluded that children with higher PA in childhood had thinner retinal venular caliber. Our results suggest that PA during childhood positively impacts the retinal microcirculation and that retinal vascular analysis may be a possible assessment to detect microvascular impairments in children with an increased risk of future cardiovascular disease.

Physical activity and myopia in Danish children-The CHAMPS Eye Study

PURPOSE

To determine associations between physical activity (PA) and myopia in Danish school children and investigate the prevalence of myopia.

METHODS

This is a prospective study with longitudinal data on PA in a Danish child cohort. Physical activity (PA) was measured objectively by repeated ActiGraph accelerometer measurement four times with different intervals (1-2.5 years) at the mean ages 9.7, 11.0, 12.9 and 15.4 years. Mean intensity of PA was estimated as counts/minutes, and time spent in sedentary, light, moderate and vigorous PA was summed using defined cut-off points. The ophthalmologic examination was conducted at the mean age of 15.4 ± 0.7 years and included cycloplegic autorefraction and biometry.

RESULTS

A total of 307 children participated in the Childhood Health, Activity, and Motor Performance School (CHAMPS) Eye Study. The cycloplegic spherical equivalent (SE) was 0.30 ± 1.46 dioptres. The prevalence of myopia was 17.9% (SE ≤-0.5 dioptres). Mean axial length (AL) was 23.5 ± 0.9 mm. For all participants, the overall mean daily distribution of PA was 67.2% in sedentary, 25.6% in light, 4.4% in moderate and 2.9% in vigorous PA. Age- and sex-adjusted linear regression showed no association between PA and SE or AL. In a prospective slope analysis, there was no association between accumulated PA during the 7 years and AL or SE.

CONCLUSION

The prevalence of myopia among Danish children was 17.9%. By logistic regression and slope analysis, we found no association between PA and myopia, in this first of its kind study based on objective and repeated PA data.

Physical activity in relation to development and progression of myopia - a systematic review

On a global scale, myopia is one of the most common causes of visual impairment. Given the increasing prevalence of myopia, it is vital to understand the pathogenesis and to identify potential interventions. Some studies have described physical activity as a potential correlation for myopia. The objective of this study was to make a systematic review regarding the correlation between physical activity and myopia. A total of 263 papers were identified in a systematic database search of PubMed/Medline and Embase. Five steps of screening removed studies of a low evidence quality and animal studies. Studies included had refractive error and physical activity (as measured by questionnaires, accelerometers and cycle ergometers) as separate, well-defined outcomes. Nine studies (six cross-sectional, two cohorts and one case-control study) with a total of 17 634 subjects were included. Six studies demonstrated a reverse association between physical activity and myopia. Three studies supported this, but also attributed the results to time spent outdoors and not physical activity per se. One cross-sectional study found no relation. We could not identify trends among the papers regarding the type of studies, population sizes, ethnicity or age of study subjects. A consistent relationship between more physical activity and less myopia was observed. No evidence of physical activity as an independent risk factor for myopia was seen. Evidence suggests that time outdoors remain the most important factor. Future studies should include objective measurements of physical activity to determine a potential independent effect. Distinction between physical activity and outdoor exposure remains important.