Abstract
Purpose :
The major risk factors for diabetic retinopathy are chronic hyperglycemia and the duration of diabetes mellitus. Long-term blood glucose level fluctuations are known to affect diabetic microvascular complications such as diabetic nephropathy. On the other hand, relationship between long-term blood glucose variability and the occurrence of diabetic retinopathy is controversial. The purpose of this study was to investigate the relationship between the HbA1c coefficient of variation and the incidence of diabetic retinopathy and to investigate the effect of blood glucose variability on the development and progression of diabetic retinopathy.
Methods :
We retrospectively studied 434 type 2 diabetes patients who underwent retinopathy examination and hemoglobin test. Patients with diabetic retinopathy at baseline were excluded from the study. Gender, age, duration of diabetes, HbA1c, Hb and cholesterol were investigated through a chart review. Univariate and multivariate analyzes were performed on factors affecting the progression of diabetic retinopathy. Based on the median value of the HbA1c and the median of the HbA1c-CV, we divided the groups into four groups and examined Kaplan-Meier analysis to see if there was any difference in the progression of diabetic retinopathy between groups according to time.
Results :
One-way ANOVA showed that the HbA1c coefficient of variation was higher in the group with moderate NPDR progression than in the group with no DR or with mild NPDR group. Kaplan-Meier analysis showed that the higher HbA1c-CV groups were, the higher the risk of progression of diabetic retinopathy. In the multiple cox regression analysis, HbA1c-CV was independently correlated with the progression of diabetic retinopathy when the HbA1c and other factors were corrected. ROC curve of intervitsit HbA1c difference showed relatively significant area under curve.
Conclusions :
In addition to the known risk factors for diabetic retinopathy, there is a risk that the progression of diabetic retinopathy will be accelerated if the HbA1c variability is high. Patients with 2.05 or more changes in consecutive HbA1c test have a high risk of developing diabetic retinopathy, and therefore need more careful screening for diabetic retinopathy. HbA1c and its variability have a significant positive correlation, so if glycemic control is successful, HbA1c variability may be lowered.
This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.