MP643ARE WE BUILDING TO MANY ARTERIOVENOUS FISTULA? FISTULA FIRST-THE DRAWBACKS

Abstract

INTRODUCTION AND AIMS: In hemodialysis (HD) patients, compared to central venous cateter, arteriovenous fistula (AVF) have been associated with the improvement of morbimortality related to vascular access management. Nonetheless, the survival benefit of AVF has been questioned. Our main goal was to evaluate the clinical impact and burden of building AVF in pre-dialysis patients. METHODS: We conducted a retrospective study of all patients with non-end stage renal disease referred for vascular access building between January 2014 and December 2015 in our hospital center. A total of 178 patients were included, their comorbidities, laboratorial data, doppler ultrasound evaluation and AVF complications were collected. The end of AVF follow-up was first of November of 2016 for those who didn't start HD or time of death and start of HD for those who engaged chronic HD program. RESULTS: During this time frame, of 178 AVFs placed, 87 patients remained in predialysis and 91 patients started chronic HD program. Mean follow-up time by a nephrologist was of 6.7±4.8 vs 5.4±4.8 years, p=0.06, respectively. Female patients represented 54% (n=47) and 45% (n=41) in each group (p=0.29), and mean age was 66.8±14.4 and 64.9±15.1 years-old (p=0.39), respectively. Hypertension, diabetes, peripheral arterial disease (PAD), ischemic heart disease (IHD), cerebrovascular disease (CVD) and chronic kidney disease etiology weren't significantly different between these two groups. Most patients built radiocephalic fistula (51% vs 46%), followed by brachiocephalic (39% vs 34%) and brachiobasilic (10% vs 19%) fistula and only one arteriovenous graft was placed in the HD group, p=0.28. A higher rate of thrombosis in the pre-dialysis group (26%, n=23, vs 13%, n=12, p=0.037) was observed, but secondary patency wasn't significantly different (55%, n=48 vs 67%, n=61, p=0.12). Mean number of placed AVF was higher in the pre-dialysis group (1.4±0.7 vs 1.2±0.4, p=0.006), but less procedures due to AVF complications were requested (0.2±0.5 vs 0.3±0.6, p=0.10). Mean time from AVF placement and HD initiation was 29.2±26.6 months. Twenty-one patients died without starting HD and 27 patients in HD group died by the end of the study. In a multivariate analysis including gender, age, IHD, PAD, CVD, diabetes, vascular access at the beginning of HD and antiplatelet/anticoagulation therapy, only age (HR 1.04, 95% CI 1.0-1.1, p=0.032) and functioning AVF at the start of HD (HR 0.11, 95% CI 0.04-0.63, p=0.008) were significantly associated with mortality, while IHD (HR 2.0 95% CI 1.0-3.95, p=0.047) showed a trend. Also in vascular access cumulative survival assessment, only thrombosis (HR 10.6, 95% CI 5.7-19.8, p<0.0001), IHD (HR 1.9, 95% CI 0.9-3.7, p=0.078) and arterial diameter at doppler ultrasound before access placement (HR 1.83, 95% CI 0.9-32.0, p=0.067) indicated a trend towards significance as predictors of AVF survival. CONCLUSIONS: Age, IHD and functioning AVF at the beginning of HD can be valuable predictors of mortality and useful for vascular access placement decision. Although AVF at the beginning of HD correlates to better outcome, a significant number of patients didn't benefit from it and considerable morbidity was associated to its placement.Figure 1. Kaplan-Meier survival curves in pre-dialysis and incident hemodialysis patients.