Management and outcomes of dialysis access-associated steal syndrome

Abstract

Dialysis access-associated steal syndrome (DASS) complicates arteriovenous access surgery. We describe a 10-year experience with the surgical management of DASS. DASS operations were retrospectively reviewed from July 2003 to July 2013 from a single academic institution. Demographics, symptoms, surgical details, and outcomes were collected. A total of 201 patients had 218 episodes of DASS. Mean age was 65 years, and 62% were women. DASS was caused by 175 arteriovenous fistulas (80%), 41 upper extremity prosthetic grafts (19%), and two thigh grafts (1%); 87% were brachial artery based. A portion (22%) were referred for DASS from outside practices. All patients had grade 2 (48%) or grade 3 (52%) DASS; 92% (185) were available for follow-up, with a median time to first follow-up of 23 days. Surgical procedures included ligation (73), distal revascularization with interval ligation (DRIL) (59), revision using distal inflow (RUDI) (21), banding (38), proximalization of arterial inflow (12), and distal radial artery ligation (13). There were no differences in preoperative comorbidities between treatment groups. The 30-day complications included continued steal, thrombosis, bleeding, infection, and mortality. Ligation and DRIL were performed most often for grade 3 steal. Ligation and banding were performed most acutely (median time to intervention after access creation of 39 and 24 days vs DRIL and RUDI at 97 and 100 days). Fistula preservation was 0% for ligation, 100% for DRIL, 95% for RUDI, and 89% for banding (P < .01). Improvement of symptoms ranged from 75% (banding) to 98% (DRIL) (P = .005). Women were less likely to have DRIL but more likely to have ligation (P = .001). Complications were highest in the banding (49%) and RUDI (37%) groups. Average mortality was 3.5%, with no significant differences among groups. During the study period, 3287 access procedures were performed, and access volume steadily increased (2003-2008, 1312 access creations; 2008-2013, 1975). Percentage of fistulas (79% vs 86%), incidence of steal (4% vs 6%), and percentage of DRILs (25% vs 28%) were consistent across the two study periods. DRIL and ligation were performed in patients with the most severe symptoms. Compared with ligation, DRIL has equal symptom resolution, no increase in complications, and fistula preservation. Compared with banding, DRIL resulted in superior fistula preservation and fewer complications. DRIL should be considered the preferred procedure for management of DASS in patients with a functioning autologous fistula who can tolerate a major operation.