FP716RELATIONSHIPS BETWEEN GLYCOSYLATED HAEMOGLOBIN AND DEATH AMONG DIABETIC PATIENTS UNDERGOING DIALYSIS:A NATIONAL COHORT STUDY

Abstract

Introduction and Aims: Recommended guidelines suggest that HbA1c levels be maintained below a target of 7% for diabetic patients in the general population. Due to shortened red cell survival on dialysis and the effects of uraemia on glucose control, it is unclear what relationship exists between HbA1c and mortality in end stage kidney disease. The aim of this study was to characterise the relationship between HbA1c and major outcomes in diabetic dialysis patients in order to identify optimal thresholds. Methods: We identified all diabetic patients who initiated dialysis from 1995-2008 and were followed until 2010. Data on demographic factors, comorbid conditions, and laboratory variables measured prior to the onset of dialysis were recorded for each patient. The relationship between HbA1c and mortality was explored in quintiles (Q) of HbA1c (Q1 < 5.6 %, Q2 5.6 ≤ HbA1c <6.1, Q3 6.1 ≤ HbA1c < 6.7, Q4 6.7 ≤ HbA1c <7.6, and Q5 > 7.6) with Q2 as the referent group and with a smoothed spline hazard function. Separate multivariable Cox regression models assessed relationships with mortality and hospitalisation by generating hazard ratios (HR) and 95% Confidence Intervals (CI). Results: There were 48,208 with diabetes and with recorded HbA1c values. The relationship between HbA1c and mortality followed a U-shaped association. Compared to the referent group (HbA1c 5.6-6.1, HR=1.00), mortality risks were significantly higher for diabetic patients with levels < 5.6 (HR 1.12 (1.06-1.08) and for patients with levels greater than the fourth and fifth quintiles [HR=1.10 (CI 1.04-1.15) and HR=1.20 (CI 1.14-1.26) respectively.With adjustment for case mix, these relationships were only slightly attenuated and remained significant for the lowest quintile and the highest quintile groups. The pattern of association was similar for Type 1 and Type 2 diabetes, and was not influenced by comorbidity score. (P> 0.05 for all interactions). Figure 1 illustrates the functional relationship between HbA1c and all-cause mortality using a smoothed spline hazard function. In a secondary analysis, HbA1c levels in the highest quintile were also significantly associated with elevated risk of hospitalisation from myocardial infarction, acute coronary syndrome, and stroke. Conclusions: For patients with diabetes undergoing dialysis, the association of HbA1c with mortality is not linear but u-shaped with elevated mortality risks for those with levels greater than 7.6% and less than 5.6%. The optimal values for best survival are between 5.6-6.7 % for all diabetic patients. (Figure Presented).