How much does glycated hemoglobin A1c explain the risk of diabetic retinopathy in persons with type 2 diabetes? The Diabetes Management Project (DMP)

Abstract

Purpose: : Although hyperglycemia is a major risk factor for diabetic retinopathy (DR) in persons with type 1 diabetes, the relationship of hyperglycemia and DR in persons with type 2 diabetes is less clear. This study aimed to determine the contribution of glycated hemoglobin A1c (HbA_1c) in explaining the risk of DR in persons with type 2 diabetes.

Methods: : Patients were recruited from eye clinics in Melbourne, Australia. Each participant underwent a comprehensive eye examination and completed standardized interview-administered questionnaires. Blood samples were assessed for HbA_1c; fasting blood glucose; and serum lipids. Dilated fundus photographs (two-field) were obtained and graded for the severity of DR. Commonality analysis was used to partition the regression effect of HbA_1c into unique and common effects. Unique effect identifies how much variance is unique to HbA_1c, and common effect identifies the joint effect of HbA_1c and other risk factors.

Results: : A total of 497 patients with type 2 diabetes aged between 26 and 90 years (median 65.0 years) were included in this analysis. Among these, 60% had any DR. Age, gender, HbA_1c, duration of diabetes, and insulin use were significantly associated with any DR, accounting for 45% of the total variance. The unique and common effects of HbA_1c on DR were 2.1% and 6.4%, respectively. The most importance risk factor for any DR was duration of diabetes, with a 7.9% and 8.7% of unique and common variances, respectively. Combining HbA_1c, duration of diabetes and insulin use explained a total of 38.0% variance (13% of unique variance and 25.0% of common variance). Subgroup analysis showed the effect of HbA_1c was higher in patients with PDR (unique variance: 3.0%; common variance: 9.1%).

Conclusions: : Although clinical trials and epidemiological studies suggest it is important to target hyperglycemia to prevent diabetic retinopathy, our study suggests that HbA_1c alone accounts for only a small proportion of the risk of DR. Targeting glycemic control alone is not sufficient to prevent DR in type 2 diabetes and a multifactorial risk management is needed.

Keywords: clinical (human) or epidemiologic studies: risk factor assessment • diabetic retinopathy • clinical (human) or epidemiologic studies: biostatistics/epidemiology methodology