Impact of glycemic control on survival of diabetic patients on chronic regular hemodialysis: a 7-year observational study.

OBJECTIVE To investigate the impact of glycemic control during regular hemodialysis on the survival of diabetic patients with chronic kidney disease (CKD) in a longitudinal observational study. RESEARCH DESIGN AND METHODS A total of 114 diabetic CKD patients on hemodialysis at Inoue Hospital (Suita, Japan) were surveyed from May 1995 to December 2002 (survey period 45.5 +/- 29.3 [means +/- SD] months). All subjects were categorized into three groups by mean HbA(1c) (A1C) level during the 3-month period on hemodialysis preceding entry, as follows: good (A1C <6.5%, 5.7 +/- 0.4%, n = 34), fair (6.5 or=8.0%, 9.2 +/- 0.9%, n = 41) A1C groups. RESULTS There were no significant differences in age at entry, initiation of hemodialysis, duration of hemodialysis, blood pressure, cardiothoracic ratio, serum creatinine level, or hemoglobin level among the three groups. The cumulative survival of the poor A1C group during the survey was significantly lower than that of the fair and good A1C groups as determined by Kaplan-Meier estimation (P = 0.041, log-rank test). In a multivariate Cox proportional hazard model, both poor A1C group (hazard ratio 2.889, P = 0.010) and mean A1C (1.260 per 1.0%, P = 0.003) were significant predictors of survival. CONCLUSIONS In diabetic CKD patients on regular hemodialysis, poor glycemic control is an independent predictor of prognosis. This finding indicates the importance of careful management of glycemic control even after initiation of hemodialysis.