| Objective:The primary objectives is to evaluate the association between residual renal function at the start of dialysis and survival.The secondary objectives is to find out the main risk factors affecting survival of dialysis.Methods:Incident patients with the end stage of renal disease starting dialysis in Zhong Shan Hospital, Fudan University between 1 January 2005 and 30 September 2009 were enrolled in this retrospective cohort study. Kaplan-Meier method, long-rank test, logsitic regression and Cox regression analyses were performed to examine the association between residual renal function and survival outcomes, using a follow-up through 31 March 2010. Residual renal function was evaluated using eGFR estimated by the abbreviated MDRD equation, the C-G equation and the EPI equation. The outcome inculded complications and all-cause mortality.Results:1. A total of 562 patients were included. Median follow-up time was 17 (0-58)months from start of dialysis and 141 patients died in this period. The median eGFR before dialysis was 6.75(2.90-15.08)ml/min/l.73m2,5.50(2.21-12.80)ml/min/1.73m2,4.87 (1.96-11.83) ml/min/1.73m2 calculated by the abbreviated MDRD equation, the C-G equation and the EPI equation respectively.2. Patients were classifed into four groups according to eGFR by>=10.5,8-10.5,6-8, <6 ml/min/1.73m. With eGFR declined, serum creatinine, blood urea nitrogen, serum uric acid, phosphorus, calcium and phosphate product, mean arterial pressure increased(P=0.000, P=0.000, P=0.000, P=0.000, P=0.000, P=0.018 respectively), and hemoglobin, hematocrit, proportion of male, proportion of diabetes complications, proportion of the Charlson combidity index>=5 decreased(P=0.000, P=0.000, P=0.001, P=0.000, P=0.003 respectively). The proportion of left ventricular hypertrophy complications before dialysis were 39%,41.3%,48.5%,55.2% according to four group, being increased when eGFR declined, but no significant difference between four groups.3. The death in first 3 months were more emerency dialysis(73%) and with higer CCI.The survival rates of first 3 months were 90.2%,90.2%,93%,94.3% in the groups of eGFR>=10.5,8-10.5,6-8,<6 ml/min/1.73m2, increased with eGFR declined but there was no significant difference and no relationship between the first 3 month survival and eGFR according to logistic regression results.Without patients died in the first 3 months, the survival rates of 1 year were 100%,100%,93.2%, 90.3%, declined while eGFR declined (P=0.022). The multivariate Cox regression model indicated age, cerebrovascular accident, the eGFR in the start of dialysis(HR=0.808,95%CI=0.692-0.945, p<0.01 by the C-G equation, HR=0.791, 95%CI=0.669-0.935, p<0.01 by the abbreviated MDRD equation, HR=0.773, 95%CI=0.642-0.931, p<0.01 by the EPI equation) were the main factors to 1 year suvival. The survival rates of 2 year were 89.2%,87%,84.5%,84%, declined while eGFR declined but no significant difference and had no relations with eGFR calculated by C-G equation (P=0.056), abbreviated MDRD equation (P=0.561) and EPI equation (P=0.299).4. The first cause of death was congestive heart failure(19.1%), the second was infection(18.4%) and the third was cerebrovascluar accident(15.6%). There was no difference of overall mortality rates between four groups using Kaplan-Meire survival curve. The conclusion were same in the subgroup of peritoneal dialysis, chronic glomerulonephritis as primary renal disease, diabetic nephropathy as primary renal disease. In the patients treated by hemodialysis, mortality was higher in group with eGFR 8-10.5 ml/min/1.73m2 (p=0.033 vs group with eGFR6-8ml/min/1.73m2, p=0.005 vs with eGFR<6ml/min/1.73m2)5. The multivariate Cox regression model indicated that there was no significant difference of survival in level of eGFR when adjusted by other case-mix differences and age, CCI, cerebrovascular accident, albumin, dialysis modility, emergent dialysis because of other reasons were the main factors to survival of dialysis.6. In the subgroup of chronic glomerulonephritis as primary renal disease(including hemodialysis and peritoneal dialysis), the eGFR in the start of dialysis was the benefit factor, with death risk decreased 16.6%,15%,18% when eGFR increased 1 ml/min/1.73m2 calculated by the abbreviated MDRD equation, the C-G equation and the EPI equation respectively (HR=0.834,95%CI=0.736-0.946, p<0.01; HR=0.850, 95%CI=0.758-0.953, p<0.01; HR=0.820,95%CI=0.715-0.940, p<0.01 by the abbreviated MDRD equation, the C-G equation and the EPI equation respectively) to survival. In the subgroup of Chronic glomerulonephritis treated by peritoneal dialysis, the death risk were decrease 33%,29%,35% with eGFR increase 1 ml/min/1.73m2 by the abbreviated MDRD equation, the C-G equation and the EPI equation respectively. In the subgroup of peritoneal dialysis, the eGFR in the start of dialysis was also benefit to survival, with HR=0.90,95%CI为0.81-0.99, p<0.05; HR=0.89, 95%CI=0.81-0.98, p<0.05; HR=0.86,95%CI=0.77-0.97, p<0.05 by the abbreviated MDRD equation, the C-G equation and the EPI equation respectively.7. No relationship were showed between survival and eGFR in the subgroup of hemodialysis and diabetic nephropathy as primary renal disease. The mean levels of eGFR between the death and the survivors were no difference in the subgroup of chronic glomerulonephritis treated by hemodialysis, the subgroup of diabetic nephrology, the subgroup of elective dialysis.Conclusion:1. The eGFR before dialysis in this study was much lower than the national dialysis gindlines suggestions.2. More and even serious complications occured when eGFR in the start of dialysis declined including more left ventricular hypertrophy, more calcium and phosphorus metabolism disorder, more anemia, higher blood pressure and higher serum uric acid.3. The higher mortality in first 3 months in higher eGFR group may resulted from higher portion of emergency dialysis. Initiation dialysis early maybe benefit to survival in ESRD patients choosing perioneal dialysis and whose primary renal disease was chronic glomerulonephritis treated by perioneal dialysis. However in total dialysis patients, hemodialysis patients including chronic glomerulonephritis patients, diabetic nephropathy as primary renal disease, there was no significant differentence. |
PDF Full Text Request