1. ¹Renal Unit, Joslin Diabetes Center, One Joslin Place, Boston, Massachusetts, USA
2. ²Fresenius Medical Care, Lexington, Massachusetts, USA

Correspondence: ME Williams, Renal Unit, Joslin Diabetes Center, One Joslin Place, Boston Massachusetts, USA. E-mail: mark.williams@joslin.harvard.edu

Received 16 March 2006; Revised 20 June 2006; Accepted 27 June 2006; Published online 30 August 2006.

Abstract

Diabetes mellitus (DM) constitutes a major end-stage renal disease (ESRD) health problem. Glycemic control is fundamental to the management of diabetes and its complications, and relies on monitoring of hyperglycemia. We therefore performed a primary data analysis of glycemic control and survival on a large national ESRD database. Ninety-five percent of patients with DM had type II diabetes (N=23 504), and five percent had type I diabetes (N=1 371). For the combined population, the mean hemoglobin A1c (HgbA1c) was 6.77%, and the mean random blood glucose was 168 mg/dl. Mean HgbA1c values were >7.0% in 35% and >8.5% in 14%. Mean HgbA1c values were below 5% in 11.3% of patients. Type I study patients tended to have higher HgbA1c values. Most patients (75.8%) had three or more random blood glucose determinations within 90 days preceding the HgbA1c measurement. The HgbA1c showed only a weak correlation with mean random glucose values (R² 0.3716; s.e.=1.36). The survival rates in the subsequent 12-month period ranged from 80 to 85% across different HgbA1c strata. Kaplan–Meier survival curves grouped by HgbA1c levels showed no correlation between HgbA1c and survival at 12 months. More studies are needed to refine recommendations for the role of HgbA1c and glycemic control in this patient population.