Age at diagnosis, glycaemic control and the development of retinopathy in a population-based cohort of Type 1 diabetic subjects in Canterbury, New Zealand.

The aim was to determine the relationship between age at diagnosis, glycaemic control and the development of retinopathy in a population-based cohort of Type 1 diabetic subjects. At 1 January 1984, there were 286 individuals with Type 1 diabetes (and age of onset<20 years) on the Canterbury, New Zealand population register who had at least 2 prospective HbA(1c) readings (from 1 January 1984). Of these, 107 already had retinopathy. Of the 179 subjects without retinopathy at baseline 63 developed retinopathy during follow-up. After controlling for duration of diabetes (in the whole group), age at diagnosis was found to be a significant predictor of HbA(1c) level (P=0.001), with higher (mean+/-SD) baseline HbA(1c) in the 10-14 age group (7.95+/-2.14%), compared with (7.62+/-1.77%) in the <10 year group and (7.39+/-2.57%) in the >14 year group. The major predictors of retinopathy (in those without retinopathy at baseline), however were duration of diabetes (mean time to development of retinopathy decreases by 14% (95% CI 10-17%) for each year), baseline HbA(1c) (for each unit increase, mean time to development of retinopathy decreased by 23% (95%CI 13-32%) and HbA(1c) slope (average annual change). Peri-pubertal age at diagnosis (10-14 years) did not influence the time to onset of retinopathy over and above that attributed to duration of diabetes and glycaemic control.