Factors predicting survival during replacement therapy for end stage renal disease

Background: The number of patients who are started on RRT is constantly increasing, thus, it is essential to know what life expectancy is being offered, and what factors predict survival. Material and methods: Retrospective study of a sample including over 80% of the patients (468) who were started on RRT in our region between 1985 and 1994, in order to find factors predicting survival, analysing global and univariate (Kaplan-Meier) survival, as well as relative death risks using Cox multivariate method. Results and conclusions: In our sample, global survival at 12, 24, 60 and 96 months was 90% 82%, 61% and 50%, respectively. Mean survival time was 80 months, with 95% confidence limits (Cl 95) of 75 and 85 months. When comparing survival curves, differences significant were observed for the following: age, main, diagnosis functional status, previous renal transplant, haemoglobin level, serum creatinine bevel and comorbidity index. Univariate analysis did not show significant differences in survival time of RRT patients with respect to: year of starting therapy, sex, marital status, haematocrit or serum urea level. When the multivariate model was applied, it was observed that death risk was reduced by variable "having received a transplant" (RR = 0.89; CI 95: 0.85-0.94). Death risk was increased by variables "main diagnosis = diabetes" (RR = 2.26; CI 95: 1.35-3.79) and "functional status grades 1 and 2" (RR = 5.13; CI 95: 2.7-0.74 and RR = 1.46 CI 95: 0.98=2.17 respectively). The remaining variables identified by the univariate model were longer significantly associated with survival.