SP380CHARACTERISTICS OF LONG-TERM SURVIVORS AMONG DIABETIC HEMODIALYSIS PATIENTS (DATA FROM THE 4D STUDY POST-TRIAL FOLLOW-UP)

Abstract

Introduction and Aims: Characterization of long‐term survivors among patients with type 2 diabetes mellitus (T2DM) on hemodialysis Treatment. Methods: Data were used from the German Diabetes Dialysis Study (4D study), a multicenter, prospective, randomized trial among 1255 hemodialysis patients with T2DM, and its post‐trial observational follow‐up. Patients' survival was analyzed calculating hazard ratios for ten parameters at baseline, chosen by clinical expertise, literature and obtainability in clinical practice. Furthermore, subgroup analyses by age (above and below 66 years [age median]) and sex were performed. Results: Median follow‐up was 11.5 years. Of 1255 patients 103 survived the in‐trial and post‐trial period. Survival analyses showed that diabetic women on dialysis lose their usual survival advantage (HR: 1.019, 95% CI: 0.890, 1.165, p = 0.789). Higher age (HR: 1.031, 95% CI: 1.022, 1.040, p < 0.001) and need of care (HR: 1.199, 95% CI: 1.031, 1.394, p = 0.018) were associated with higher mortality as were cardiovascular (HR: 1.423, 95% CI: 1.249, 1.621, p < 0.001) and peripheral vascular co‐morbidities (HR: 1.549, 95% CI: 1.364, 1.759, p < 0.001) and higher HbA1c (HR: 1.084, 95% CI: 1.032, 1.139, p = 0.001). However, higher albumin (HR: 0.723, 95% CI: 0.589, 0.887, p = 0.002) and BMI (0.977 (95% CI: 0.964, 0.992, p = 0.002) seem to be protective. Cerebrovascular diseases (HR: 0.858, 95% CI: 0.729, 1.009, p = 0.063) and diabetes duration (HR: 0.998, 95% CI: 0.991, 1.006, p = 0.685) did not have a significant effect on all‐cause mortality. Subgroup analyses by age (figure 1) showed increased mortality for patients under 66 with higher HbA1c, cardiovascular diseases and need of care and for older patients with lower BMI and albumin. Subgroup analyses by sex (Figure 2) revealed that mortality in both sexes was basically influenced by the same risk factors. Yet in women need of care and decrease in BMI or albumin had a stronger influence on mortality, whilst in men cardiovascular and peripheral vascular diseases had a stronger effect. Conclusions: The average long‐term survivor can be described as follows: he (!) is relatively young, is self‐dependent, has little co‐morbidities and a low HbA1c. Subgroup analysis identified patients necessitating further individualized criteria for prognosis estimation and therapy, which should strongly depend on age and sex. (Figure presented).