Abstract

To study the progression of retinopathy in patients with mild diabetes mellitus, we examined, in a cohort study, 347 patients treated with diet alone at baseline. The patients participated in an ophthalmological screening and control programme, and diet-treated patients who were examined between January 1990 and July 1992 were included in the study and followed until October 1995. Mean follow-up was 3.4 +/- 1.1 years. The alternative classification of the Wisconsin study was used to classify retinopathy, and the mean HbA1c values for the study period, to estimate the level of glycaemic control. At baseline, 314 of the patients (90.5%) had no retinopathy, and 33 (9.5%) had mild non-proliferative diabetic retinopathy. In 296 patients there was no retinopathy progression, in 27 patients there was progression by 1 level in the retinopathy scale, and in 24 patients by 2 levels or more. In 2 patients there was progression to proliferative diabetic retinopathy. The mean HbA1c (%) was 6.5 +/- 1.3. Higher HbA1c correlated to increased progression (r = 0.16; p = 0.005), and in a multivariate analysis, HbA1c remained associated with a progression of retinopathy by 2 levels or more, with a relative risk of 1.4 per percent increase in HbA1c (95% CI 1.1-2.0; p = 0.02). Furthermore, the presence of any retinopathy at baseline was associated with progression with a relative risk of 1.7 (95% confidence interval 1.1-2.8; (p = 0.02). These data indicate that even slightly elevated levels of HbA1c might be associated with a risk of retinopathy progression.