ƩexcessA1C index, the sum of yearly excess HbA1c values during the total period of diabetes, may have the potential to predict retinopathy by a linear regression setting regardless of duration in type 1 diabetes: a subgroup analysis of DCCT/EDIC data.

Abstract

Aims: To find an index of glycemic exposure that predicts retinopathy by a simple regression setting regardless of duration in type 1 diabetes which might be useful for the care of diabetes.

Materials and methods: To exclude the possible disturbing effect of metabolic memory, we examined a subgroup of patients with glycohemoglobin A1c (A1C) data for the total period of type 1 diabetes selected from the Diabetes Control and Complications Trial/Epidemiology of Diabetes Interventions and Complications data. Three indices-(1) mean value of yearly A1C (mA1C), (2) sum of yearly A1C values (ƩA1C), and (3) sum of yearly A1C values above 6.5% (ƩexcessA1C)-were assessed as potential candidates. Development of retinopathy was defined by ≥ 3-steps' progression of retinopathy from baseline.

Results: The areas under the receiver operating characteristics curves of the indices for development of retinopathy at years 5, 9, and 13 after the onset of diabetes were the same: 0.8481, 0.8762, and 0.8213, respectively, indicating that each index was substantially capable of predicting development of retinopathy at each timepoint. Linear regression analyses showed that each index had significant and substantial linear relations to retinopathy at each timepoint: all P < 0.0001 for slopes; contribution rate R² = 0.21 (year 5), 0.46 (year 9), and 0.48 (year 13) for each index. But only ƩexcessA1C index appeared to have similar linear relations to retinopathy at all three timepoints (interactions by timepoint: for slopes: P = 0.1393; for intercepts: P = 0.9366).

Conclusion: ƩexcessA1C may have the potential to predict retinopathy by just one linear regression setting regardless of duration in type 1 diabetes.