SP591PATIENT CHARACTERISTICS AND OUTCOME OF HAEMODIALYSIS VASCULAR ACCESS PATENCY - A SINGLE CENTRE EXPERIENCE

Abstract

INTRODUCTION AND AIMS: Arteriovenous fistulae (AVF) and arteriovenous grafts (AVG) are key to delivering haemodialysis (HD). Eventually, these vascular accesses (VA) will experience progressive deterioration in function, requiring endovascular or surgical intervention. We aim to identify factors that could influence the functioning of VA. METHODS: We retrospectively identified patients who underwent VA formation at our hospital from October 2009 to December 2015. Demographics, background renal disease and comorbidities were collected from our renal database. We analysed association between age, sex and comorbidities (diabetes (DM) and hypertension (HT)). In addition, we identified the anatomical location of AVF, referral to one stop vascular access clinic (NEPVA), time of angioplasty, number of angioplasties, steal syndrome and date of failure. In NEPVA clinic, the patient receives multi-disciplinary (MDT) assessment by the operating surgeon, nephrologist and specialist nurse including US Doppler with follow-up 6 weeks post VA formation. In contrast, patients who were not referred to NEPVA received an US Doppler with neither MDT assessment nor follow-up. RESULTS: 427 patients (260 M, 167 F) underwent VA formation. Mean sample age was 63 years. 168 patients had DM (39.3%) and 247 had HT (57.8%). Causes of endstage renal failure were: uncertain aetiology (41.0%), DM nephropathy (30.4%), glomerulonephritis (21.6%), polycystic kidney disease (3.75%) and HT nephropathy (3.28%). Stage of chronic kidney disease (CKD) when VA was formed were predialysis, CKD 4/5 (54.8%), prevalent HD (37.9%), prevalent peritoneal dialysis (6.09%) and transplant (1.17%). 538 VA were formed (209 radiocephalic AVF, 229 brachiocephalic AVF, 87 brachiobasilic AVF and 13 AVG). Proportion of patients who had 1, 2, 3, 4 and 5 VA formed was 338, 73, 11, 4 and 1 respectively. 203 patients (47.5%) were seen in the NEPVA clinic while the remainder (224) went directly to VA formation. 455 VA were usable for dialysis, giving a primary failure rate of 15.4%. (11.8% NEPVA, 18.9% no NEPVA). Secondary revision rate was 8.57%. Patency rate at 12 months was 58.2%. Median time to angioplasty was 379 days. Angioplasty was performed in 278 VA (51.7%) and percentage of angioplasties performed once, 2 times, 3 to 5 times, 6-10 times and more than 10 times are 19.3%, 11.3%, 14.3%, 5.8% and 0.6% respectively. Incidence of steal syndrome with critical ischaemia was 3%. Number of VA that failed was 74 (16.3%). Median time to failure was 335 days (s.d. 526). Failure occurred in AVG (23.1%) followed by brachiobasilic AVF (21.8%), radiocephalic AVF (13.9%) and brachiocephalic AVF (10.0%). We found no correlation between age or having DM contributing to VA failure. However we found that being male, having HT, type of VA and having a prior angioplasty was associated with higher rates of VA failure; (p values 0.014, 0.008, 0.032 and 0.001 respectively). Being assessed in NEPVA clinic was associated with lower rates of primary VA failure (p value 0.022). Meanwhile, revision of VA was not associated with higher rates of failure (p value 0.744). CONCLUSIONS: 1. NEPVA assessment is associated with a lower primary failure rate. In addition, higher radiocephalic AVF and lower brachiobasilic AVF formation rates were observed. 2. Good blood pressure control and location of VA in brachiocephalic and radiocephalic sites are associated with longer VA patency. 3. Close monitoring of VA function during the first year of use is recommended as we found in our study, most complications occurred during this period.