Glycated haemoglobin and the incidence of end-stage renal disease in diabetics.

BACKGROUND The relationship between glycated haemoglobin and the incidence of end-stage renal disease (ESRD) in patients with diabetes remains uncertain, especially in those with decreased glomerular filtration rate (GFR). The aim of this study was to assess the appropriate HbA(1c) level for diabetics for minimizing the incidence of ESRD and all-cause mortality. METHODS A cohort of patients aged 25 years or older who had been treated for diabetes was generated from the Seoul National University Bundang Hospital database using diagnosis code and prescribed medication during 2004. The 4474 patients were classified into three groups according to the baseline HbA(1c) in 2004 (HbA(1c) < 6.50%, 6.50-7.49% and ≥ 7.50%; termed groups 1, 2 and 3, respectively). The outcomes were extracted from the database of Statistics Korea for mortality and registry in the Korean Society of Nephrology for ESRD incidence. RESULTS Ninety patients developed ESRD during 5.29 ± 1.22 years of mean follow-up period. Group 1 patients showed the lowest incidence of ESRD (P = 0.003). Compared with this group, the adjusted hazard ratio of ESRD was 2.915 and 4.219 in groups 2 and 3, respectively. The incidence of ESRD increased in patients with HbA(1c) ≥ 6.50% compared with the patients with HbA(1c) < 6.50%, regardless of GFR. However, HbA(1c) < 6.50% showed no benefit on ESRD development in patients older than 80 years and in patients with diabetic duration > 10 years. All-cause mortality was not associated with the level of HbA(1c). CONCLUSIONS HbA(1c) < 6.50% was associated with reduced development of ESRD in all patients and later stages of chronic kidney disease.