Impact of HbA1c absolute change patterns on diabetic retinopathy risk in type 2 diabetes

To explore the effect of absolute changes in glycated hemoglobin A1c (HbA1c) on the risk of diabetic retinopathy (DR) in individuals with type 2 diabetes (T2D).

This retrospective cohort study involved 3,767 people living with type 2 diabetes without DR, managed at six clinics in Taiwan from January 1, 2002, to April 11, 2018. Demographic and laboratory data were collected regularly. Using the K-means clustering algorithm, participants were grouped into four categories based on HbA1c absolute changes: stable, increasing, slightly decreasing, and greatly decreasing. The correlation between HbA1c change patterns and DR risk was analyzed using the Cox proportional hazards regression model.

Four distinct HbA1c change patterns were identified, with a mean follow-up period of 9.5 ± 3.4 years. After adjusting for confounding variables, the Cox model indicated higher DR risks for the increasing, slightly decreasing, and greatly decreasing groups compared to the stable group. However, after adjusting for baseline HbA1c, there was no significant difference in DR risk between the slightly decreasing group (HR = 0.9, CI: [0.7, 1.0], p = 0.1) and the stable group. The increasing group had a significantly higher risk (HR = 1.6, CI: [1.4, 2.0], p < 0.001), while the greatly decreasing group had a significantly lower risk (HR = 0.6, CI: [0.4, 0.8], p < 0.001) than the stable group.

While HbA1c variability is a risk factor for DR, a substantial reduction in HbA1c is protective for people living with type 2 diabetes with very high baseline HbA1c.