We are grateful to Śnit et. al. (1) both for their interest in our study and for their reexamination of our findings in their dialysis patients. We demonstrated that diabetic patients on regular hemodialysis with poor glycemic control A1C (>8.0%) have poor prognosis of life and that A1C was an independent predictor after adjusting multivariate risk factors (2). In response to our findings, they have commented that there were no significant differences in the cumulative survival rate of diabetic patients on dialysis among their three groups according to the categorization by A1C, as in our study. There may be a few reasons for the discrepancy in our findings. First, their subjects were heterogeneous in type of diabetes and mode of dialysis. Patients with type 1 diabetes usually have a lower risk of cardiovascular disease, a major cause of death in dialysis patients, compared with type 2 diabetic patients, and approximately one-third of their subjects had type 1 diabetes. Furthermore, one-third of their subjects underwent peritoneal dialysis. In general, peritoneal dialysis patients have considerably different metabolic disorders and risk factors for cardiovascular diseases compared with hemodialysis patients. The use of dialysate with high-glucose solution and continuous ultrafiltration induces different states of glycemic control, dyslipidemia, nutrition, and cardiac function. Furthermore, their patients with poor glycemic control and complications were prescribed higher doses of dialysis, which may result in a better prognosis. Second, as Śnit et al. also pointed out, age and creatinine level of their subjects differed considerably from our subjects and among their three groups. Third, the existence of cardiac diseases and medications among their groups were not clearly commented, although no significant difference in left ventricular mass was described. In our study, these factors also strongly affected our findings.
Many confounding factors contribute to the prognosis of life in diabetic patients on dialysis. Thus, more careful analyses and interpretations will be needed in such an observational study. We hope that clinical implications of glycemic control in dialysis patients will be reexamined by many investigators for the sake of a better life for dialysis patients.
