Key points are not available for this paper at this time.
The development of proliferative diabetic retinopathy was studied in three cohorts consisting of 292 patients with recent juvenile-onset, type I (insulin-dependent) diabetes who were followed 20-40 yr beginning in 1939, 1949, and 1959. The risk of this severe eye complication was almost nonexistent during the first 10 yr of diabetes, rose abruptly to its maximum level (approximately 30/100 person-years), and remained at that level for the next 25 yr. This pattern did not vary with sex, age at onset of diabetes, or level of glycemic control during the first 5 yr of diabetes. However, the risk of proliferative retinopathy was strongly related to the level of glycemic control during the several years preceding onset of this complication. This was a dose-dependent relationship, with patients in the highest quartile of the distribution of the index of frequency of hyperglycemia having a 10-fold higher risk than individuals in the lowest quartile. A virtually identical pattern was observed in patients who developed diabetes in 1959 as was observed in those who developed diabetes in 1949 or 1939. In contrast, diabetic nephropathy as evidenced by persistent proteinuria showed a lower incidence in the 1959 than in the 1939 cohort. In conclusion, these incidence data do not support the notion that the risk of proliferative retinopathy is mainly a function of duration of diabetes. Instead, the pattern of occurrence of this severe eye complication in type I diabetes suggests that the process leading to the development of proliferative retinopathy consists of two or more stages and that progression through each stage may be governed by different factors.
Królewski et al. (Mon,) studied this question.