Objectives: To investigate the distinct effects of variation in HbA1c and mean HbA1c on diabetes-related complications in type 2 diabetes patients.
Methods: In a Diabetes Shared Care Program, 1869 type 2 diabetes patients were followed up for an average of 9.5 years. Mean HbA1c (HbA1cmean) and standard deviation of HbA1c (HbA1c-SD) were determined during the initial 5 years. Nephropathy (urine albumin-to-creatinine ratio [UACR] greater than 300 mg/g and doubling of serum creatinine), any and advanced retinopathy, and all-cause and cardiovascular disease mortality were among the clinical outcomes.
Results: UACR > 300 mg/g (Hazard ratio [HR] 1.308 [95% confidence interval [CI], 1.194–1.433]), any retinopathy (HR 1.274 [1.171–1.385]), and advanced retinopathy (HR 1.237 [1.014–1.509]) were all independently associated with HbA1c-mean. UACR > 300 mg/g, doubling of serum creatinine (HR 2.133 [1.470–3.095]), all-cause mortality (HR 1.880 [1.561– 2.266]), and CVD mortality (HR 1.431 [1.069–1.915]) were all independently associated with HbA1c-SD. HbA1c-mean was more correlated with any retinopathy than HbA1c-SD, whereas HbA1c-SD was more correlated with all-cause and CVD mortality and doubling of serum creatinine.
Conclusion: Majority of diabetes-related complications in type 2 diabetes patients were predicted by HbA1c-mean and HbA1c-SD. In any case, HbA1c-mean was more powerful at anticipating retinopathy, while HbA1c- SD was more successful at foreseeing disintegration of renal capability and expanded mortality.
