In their Diabetes Care article, Oomichi et al. (1) stated that poor glycemic control is an independent predictor of poor prognosis of survival in patients on chronic regular dialysis. In the dialysis center of the Clinic of Internal Medicine, Diabetology and Nephrology in Zabrze, we have undertaken regular dialysis of ∼200 patients. We have recently published our experiences in regards to renal replacement therapy, especially concerning diabetic patients (2,3), but we have never analyzed the relationship between glycemic control and survival of dialyzed patients. During the years 1992–2003, we dialyzed a total of 409 patients, 240 (58.7%) of which had diabetes with a mean ± SD age of 54.0 ± 12.7 years. Of the 240 diabetic patients, 80 had type 1 diabetes and 160 had type 2 diabetes. In addition, 167 diabetic patients (69.6%) were on hemodialysis and 73 (30.4%) were on peritoneal dialysis, either continuous ambulatory peritonial dialysis or atrial premature beat. The mean A1C level was 7.7 ± 1.4%.
We divided patients into three groups, as in the original study by Oomichi et al.: group I, A1C <6.5%; group II, A1C 6.5–7.9%; and group III, A1C >7.9%. Mean creatinine level (660.2 ± 255.1 vs. 574.8 ± 253.0 μmol/l) and mean age (55.3 ± 12.4 vs. 51.5 ± 13.1 years) at the start of renal replacement therapy were significantly higher in group I compared with group III (P < 0.05 and P < 0.02, respectively). Other parameters significant to patients’ survival, such as blood hemoglobin concentration, index of dialysis on dialysis, duration of renal replacement therapy, left ventricle mass index, and blood pressure, were similar and statistically significant in all groups. Cumulative survival rate of diabetic patients undergoing dialysis, depending on glycemic control measured by A1C concentration, was not statistically different in all groups.
One explanation for the discrepancy between our observations and those of Oomichi et al. is the fact that our group of patients with good glycemic control undergoing renal replacement therapy was older and had higher mean blood creatinine levels. A second reason for the discrepancy is the higher prescribed dose of dialysis, which is normal procedure in our center in patients with complications and poor glycemic control. We conclude that it was not only A1C, which reflected the state of accurate glycemic control, that influenced the state and prognosis of dialyzed patients; other factors must be taken into account.
