A large observational study of incident and prevalent haemodialysis patients from Canada showed similar findings.8 Two cohorts Ku-0059436 of patients, those with diabetes and those without, were created between 1994 and 2000 and followed until 2001. Diabetic patients had significantly higher comorbidities and not surprisingly, once on dialysis, diabetic patients had lower rates of survival

than non-diabetics (3-year survival 55% vs 68%, P < 0.0001). This finding was consistent with that reported by the Canadian Organ Replacement Register, which reported a 3-year survival of 52% for diabetics and 65% for non-diabetics.9 A retrospective analysis of 750 Spanish peritoneal dialysis patients was published in 2002.10 This group analysed comorbidity and mortality in type 1 diabetics, type 2 diabetics and non-diabetic patients. Different comorbidity factors such as age and the presence

of CVD at the initiation of peritoneal dialysis were analysed as well as the incidence of peritonitis, need for hospitalization and among other factors, mortality rate. The number of comorbid conditions when starting Staurosporine cell line the treatment (comorbidity index) and the peritonitis incidence was higher for type 2 diabetics and death during the first year of treatment was higher for type 1 diabetics. The actuarial survival curves showed a higher mortality for type 2 diabetics with no differences between non-diabetics and type 1 diabetics after adjustment for age. The mortality odds ratio

was 1.78 for type 2 diabetics and 1.13 for type 1 diabetics, differences that were not significant after age at >70 years and CVD were added to the variables analysed. This study thus highlighted that while cardiovascular comorbidity was responsible for the higher mortality found in the first year in type 1 diabetics compared with Adenosine triphosphate non-diabetics, both age and CVD were responsible for the higher mortality and complications faced by the type 2 diabetics. Infection is another leading cause of death in diabetic patients receiving haemodialysis, and septicaemia has been reported to be responsible for 75% of deaths related to infections.11 The infected dialysis access or infected foot, impaired cellular immunity and humoral immunity and nutritional deficiency may play major roles. Very few studies have examined the association of glycaemic control (HbA1C) and clinical outcomes in the dialysis population.12 Four of these studies12–14,16 had small sample sizes of less than 150 subjects and four were performed in exclusively Asian populations.12,13,16,17 The three largest studies15,17,18 have conflicting results. Williams et al.15 performed a primary data analysis of glycaemic control and survival on 23 504 diabetic dialysis patients in the USA. Five per cent of the population had type 1 diabetes and patients were followed for 12 months. No difference in survival was observed across the different HbA1C strata with survival rates ranging from 80% to 85%.