Prophylactic Closing Of Av Fistula Prevents High Output Heart Failure - A Single-Center Randomised Controlled Trial

Abstract Background and Aims AV fistulas are the favoured way ensuring a safe long-term access for patients with need for hemodialysis. Various studies, including a prior retrospective trial at our department, showed that AV fistulas are an additional burden on the cardiac-vascular system leading to frequent fistula closure in post-transplant patient. In literature, there is a widespread approach in dealing with high-flow fistulas in post-transplant patients with stable kidney function is reported. Method For this trial, patients with a stable graft function (min. 3 months post-transplant) and a brachial artery flow of min. 1,500 ml/min were recruited. Initially planned for multi-centric analysis, 28 patients were randomised in Innsbruck. Exclusion criteria were a pre-existing cardiac failure with ejection fraction less than 25%, NT-pro-BNP > 1,400 ng/l and NYHA score ≥ III. These patients were randomised in a 1:1 ratio in intervention group with immediate fistula ligature after randomisation, and control group with no surgical intervention. Patients in the control group had fistula ligature only if at least two criteria for ligature occurred. These criteria were selected in agreement with of the department for cardiology and encompassed dilation of right atrium, of inferior vena cava or of pulmonary artery, left-ventricular eccentricity index < 1 or a systolic pulmonary artery pressure of > 35 mmHg. The main endpoint high-output heart failure was defined by min. 2 echocardiographic criteria and signs of congestive heart failure such as worsening in NYHA score. A follow-up procedure of 24 months with quarterly measurements of kidney function, NT-proBNP and lactat dehydrogenase as well as a biannual echocardiographic check-up were performed. Statistics included paired t-test and Wilcoxon-rank test for median/mean comparison and chi-quadrat test for absolute frequency. Results The main endpoint high-output heart failure referable to high fistula flow was reached in five of 13 control patients (38.5%), whereas in the intervention group no study patient showed a worsening in clinical and echocardiographic signs to reach criteria for ligature. Prophylactic ligature of high-flow fistulas avoided right heart failure in our patient collective (p-value 0.013, x?test). Three patients had to undergo ligature due to off-study complications such as progressive aneurysma formation and steal phenomenon. One patient with dyspnoe preferred fistula ligature but was not fulfilling echocardiographic criteria. In total nine patients in the control group were assigned to fistula ligature. Matching all conducted fistula closures median NT-proBNP was 317 ng/l pre-ligature and 223 ng/l post-ligature (p 0.003, Wilcoxon). In total, 18 of 21 patients (excluding 2 patients with lack of data) showed falling NT-proBNP values. Creatinine levels showed a decrease from 1.69 mg/l to 1.60 mg/l (p-value 0.059, paired t-test). The intervention group included 14 patients with a gender ratio of 12 men to 2 women and a mean age of 52.5 years; the same age and gender distribution was seen in the control group. Fistula flow per m?body surface did not differ significantly (980 ml/min vs. 1000 ml/min). Echocardiographic findings showed improvement in ligature patients (e.g. a drop of sPAP in 7/8 ligature patients with available data) but did not reach statistical significance. Conclusion A prophylactic ligature of high-flow AV fistulas can avoid high-output heart failure and a more liberal approach to closure AV fistulas may be discussed in future.