Abstract: Objective  To further explore whether the management of chronic kidney disease (CKD) before dialysis has a long-term effect in the period after hemodialysis, so as to contribute to the better promotion of CKD management from the perspective of economics and demography.  Methods  This study enrolled 70 patients with initial hemodialysis from the CKD Outpatient Clinic (management group, group M) and 142 patients with initial hemodialysis from the Nephrology Clinic (non-management group, group NM) in the Affiliated Hospital of Jiangsu University in the period from March 2015 to March 2018. The patients were followed up for 5 years, and the hospitalization and expenses at the initial of dialysis and within 1 year after dialysis, as well as the death at 1 year and 5 years after dialysis were compared between the two groups, and the influence of various factors on all-cause death was analyzed.  Results  When starting dialysis, the total hospitalization cost per capita in group NM was significantly higher than that in the group M (t=-3.100，P=0.002). The total hospitalization cost per capita in group M was significantly lower than that in the group NM within 1 year after dialysis (t=-2.269, P=0.024), this is mainly due to a decrease in the cost of medicines, tests, treatments, surgeries and blood transfusions. Multiple linear regression found that CKD management was independently associated with a reduction in total inpatient costs per patient, whether unadjusted, adjusted for age and sex, adjusted for comorbidities, adjusted for baseline laboratory measures at hemodialysis starting, or adjusted for vascular access at hemodialysis initiation, at the time of initiation of dialysis and within one year after initiation of dialysis (Dialysis start-up β=0.386, 0.392, 0.392, 0.359, 0.248，P＜0.001, ＜0.001, ＜0.001, ＜0.001,  =0.018, respectively; Within 1 year of dialysis β=0.151, 0.154, 0.148, 0.168, 0.343，P=0.028, 0.025, 0.038, 0.021, 0.002,  respectively). Kaplan-Meier analysis showed that the 5-year cumulative survival rate difference was statistically significant (χ2=3.947, P=0.047). The results of multivariate analysis showed that age and diabetes mellitus were independent risk factors for all-cause death (HR=1.042, 0.390,95% CI: 1.021～1.062, 0.226～0.671, P＜0.001, 0.001), while pre-dialysis CKD management was protective factor for all-cause death (HR=0.503, 95% CI: 0.295～0.857, P=0.012).  Conclusions   Pre-dialysis CKD management has a legacy effect on post-dialysis economic outcomes and can improve the long-term prognosis of maintenance hemodialysis patients, we should pay more attention to diabetes and elderly patients in CKD management.

Key words: Chronic kidney disease, Management, Hemodialysis, Costs, Mortality rate