P1463EARLY VASCULAR ACCESS CREATION IS NOT ASSOCIATED WITH BETTER OUTCOME IN PATIENTS AFTER CHRONIC DIALYSIS INITIATION

Abstract

Listen

Translate article

Abstract Background and Aims Although initiation of chronic hemodialysis with central venous catheters are associated with higher mortality compared to that with arteriovenous fistula, the association between the timing of vascular access (VA) creation and outcome after dialysis initiation is not fully investigated. Thus, we conducted single center retrospective cohort study on incident hemodialysis patients to investigate the effect of VA types and timing of VA creation on all-cause and cardiovascular mortality after dialysis initiation. Method A total of 1052 patients with accurate clinical data and outcomes was extracted from 1069 patients who started chronic hemodialysis treatment at Osaka General Medical Center between 2006 and 2015. Clinical status at dialysis initiation, all-cause and cardiovascular mortality were compared by three patient groups categorized by VA types and timing of VA creation: temporary catheter (TC) group, dialysis initiation with TC; early VA (EVA) group, dialysis was initiated more than one month after VA creation; late VA (LVA) group, dialysis was initiated within one month after VA creation. Cox proportional hazards analysis was used to examine predictive factors for outcomes. Results Median age was 70 years and 64% of patients were male. Patient number of each group was 285 (27%), 441 (42%), and 326 (31%) for TC, EVA, and LVA groups, respectively. Although there was no significant difference in eGFR at dialysis initiation among three groups ([median] 5.4 vs. 5.4 vs. 5.4 mL/min/1.73m2, respectively, P=0.881), hemoglobin and albumin were lower in TC group compared to EVA and LVA groups (hemoglobin [median] 8.3 vs. 9.5 vs. 8.9 g/dL, respectively, P<0.001; albumin [median], 3.0 vs. 3.5 vs. 3.2 g/dL, respectively, P<0.001). Kaplan-Meier curves revealed significant difference among three groups for all-cause (P<0.001) and cardiovascular mortality (P<0.001). On multivariate Cox proportional hazards analysis, timing of VA creation was not associated with all-cause and cardiovascular mortality. Although temporary catheter usage at dialysis initiation was significantly associated with all-cause mortality (HR, 1.91; 95%CI, 1.30-2.83; P=0.001), it was not associated with cardiovascular mortality (HR, 1.64; 95%CI, 0.79-3.43; P=0.186). Conclusion Timing of VA creation was not associated with all-cause and cardiovascular mortality, whereas hemodialysis initiation with TC was significantly associated with increased all-cause mortality.