RETINOPATHY

The majority of cases of retinopathy found in a recent population-based Icelandic study occurred in persons who do not have diabetes. The Icelandic study also showed a hitherto unreported association between retinopathy in non-diabetic patients and high levels of microalbuminuria, reported Fridbert Jonasson MD, at the 12th EURETINA Congress. Icelandic researchers investigated retinopathy prevalence and risk factors using data from 4,994 persons aged 67 years and older who participated in the Age, Gene/Environment Susceptibility Reykjavik (AGES-R) Study. The survey found that 516 (10.3 per cent) participants had diabetes based on a definition of an HbA1c level ≥6.5 per cent.

Based on a modified ETDRS protocol for rating fundus photographs, retinopathy was identified in 138 (27 per cent) diabetic patients and 476 (10.7 per cent) of those without diabetes. Of note, most eyes with retinopathy did not present with severe changes. The most common features were blot retinal haemorrhages and microaneurysm. Among the 138 cases of retinopathy in the diabetic cohort, there were five persons who had proliferative diabetic retinopathy and five who had clinically significant macular oedema. Among the 476 non-diabetics with retinopathy, there were only five eyes with macular oedema.

“Although the prevalence of retinopathy is about 2.5-fold higher among the diabetics than in the non-diabetic cohort, it is noteworthy that three-fourths of cases of retinopathy in this random population sample occurred among persons without diabetes,†said Dr Jonasson, professor of ophthalmology, University of Iceland, Reykjavik. Multivariate analyses were conducted to identify risk factors for retinopathy. The variables investigated included age, sex, systolic blood pressure, hypertension, HbA1c level, and microalbuminuria. Duration of diabetes and use of glucoselowering medication (insulin and oral hypoglycaemics) were also included in the statistical models for diabetic persons.

In univariate analyses, systolic blood pressure, duration of disease, and level of HbA1c were significantly associated with retinopathy among the diabetics. However, when treatments for diabetes were added into the multivariable model, disease duration dropped out, and independent predictors of diabetic retinopathy were HbA1c level (odds ratio = 1.35 for each percentage point), systolic blood pressure (odds ratio = 1.16 for each 10 mmHg), insulin use (odds ratio = 3.51) and use of oral hypoglycaemic agents (odds ratio = 1.93).

Independent risk factors for retinopathy in the nondiabetic cohort were microalbuminuria (odds ratio = 1.77) and increasing age (odds ratio = 1.30 for every 10 years). “To our knowledge, this association between retinopathy and microalbuminuria in non-diabetics is a novel finding. We believe they may both be a marker of systemic microvascular dysfunction, and we have recently identified an associated genetic variant,†Dr Jonasson said. He noted that when he sees patients with retinopathy in clinical practice, his routine has been to refer them for evaluation by a diabetologist. While this approach may be called into question by the study’s findings, developing recommendations on patient management are pending further investigation.

“More data are needed to understand the underlying pathophysiology of retinopathy in patients who are not diabetic and to establish an evidence basis for directing their clinical care,†he said. Dr Jonasson added that patients with low-grade retinopathy are also unlikely to present to the ophthalmologist with vision complaints, but that initiation of screening programmes to identify non-diabetics with retinopathy are not warranted. Further details on this study may be found in a published report [Diabetologica 2012;55:671-80].