Risk Factors and Outcomes of Revisions on Arteriovenous Hemodialysis Access

Abstract

Arteriovenous access (AV) using either a fistula (AVF) or graft (AVG) for hemodialysis (HD) access is often complicated by poor function requiring revision. We investigated the risk factors and durability of revisions on AV access sites for patients on chronic HD. This retrospective cohort study identified all patients undergoing chronic HD for over 3 months in the Champlain NephroCare network between 2003 and 2018. Access creations, revisions, and functionality were documented. Cumulative access site durability was defined as the functional duration of AV access site despite revisions. Cox regression survival analysis was performed, with competing risks for transplants and with access sites clustered by subject. We identified 1962 eligible AV access creations in 1034 subjects, composed of 1725 (88%) AVF access. Of AV, 67% underwent at least one revision, and 13% received two or more revisions. Revisions were most commonly angioplasty (95.8%). There were 354 (26%) revisions performed to assist maturation before the access site was ever functional, of which 58 were repeated revisions. Conversely, 1011 (74%) of revisions were performed to maintain functionality after the access site had been exclusively used for dialysis, of which 473 procedures were repeated revisions. Of all AV access creation attempts, 35% were never functional without secondary procedures, and another 31% experienced primary failure within the first year. However, the 12-month secondary functional patency of AV access sites was 72%. Advanced age (odds ratio [OR], 1.14; P < .01), lower arm placement (OR, 1.43; P < .01), diabetes (OR, 1.45; P < .01) and prior ipsilateral central venous catheter (OR, 1.32; P = .02) were associated with need for secondary procedures. Revisions to assist maturation were less successful on AV grafts compared with fistulas (63% vs 77%; P = .04). Revisions to maintain functionality were less successful on AV grafts, occluded access sites, or access sites with previously low dialysis flow rates. The 12-month primary patencies after revisions to maintain functionality was 82%, 73%, and 61% after the first, second, and third revisions, respectively (P < .01). Secondary procedures on AV access sites are common and generally produce accepTable Results. Multiple factors are significantly associated with the outcomes of secondary procedures. Although repeated secondary procedures produce diminishing returns, the primary patency after a third revision remains superior to that of a new AV access site.Fig 2View Large Image Figure ViewerDownload Hi-res image Download (PPT)