Arteriovenous fistulas as vascular access for hemodialysis: The preliminary experience at the University Hospital of the West Indies, Jamaica.

BACKGROUND: The demand for vascular hemodialysis access creation is steadily increasing. To satisfy the demand, a vascular access team was established at the University Hospital of the West Indies, Jamaica. The outcomes of this practice are reported.
METHODS: A retrospective study of all patients who had permanent vascular dialysis access established at the University Hospital of the West Indies between January 1, 2002, and December 31, 2006, was performed. Data were analyzed using SPSS version 12.0 (SPSS Inc, USA). A direct anastomosis between an autogenous artery and vein was considered an arteriovenous fistula (AVF). When prosthetic material was used, the access was considered to be an arteriovenous graft. Accesses that were nonfunctional after six weeks of maturation were considered to be primary failures, while those that failed after previous successful dialysis were considered to be secondary failures. Primary patency was defined as the interval between access placement and the first intervention for failure. Secondary patency was the interval between access placement and abandonment. Cumulative patency was defined as the number of accesses that remained patent over a given time period, regardless of the number of interventions performed.
RESULTS: Of 41 patients, nine were excluded due to incomplete data. Final analyses were performed on 32 patients with a mean (± SD) age of 42.3±15.3 years (range 18 to 66 years, median 43 years). The access type was an AVF in 100% of cases, which included distal radiocephalic fistulas in 27 patients, brachial-cephalic fistulas in three patients and proximal radiocephalic fistulas in two patients. Operations were performed in four (12.5%) incident and 28 (87.5%) prevalent dialysis patients. The mean delay between initiation of dialysis and AVF creation was 21.2±26.1 months (range one to 94 months, median 10 months). There were eight (25%) primary failures. Of the remaining 24 patients, there were seven (29.2%) secondary failures from thrombosis. There was primary patency for a mean of 723.9±422.1 days (range 199 to 1314 days, median 678 days). Only one (4.2%) patient had thrombectomy to prolong AVF function, resulting in secondary patency for 439 days. Cumulative patency was 62.5%, 33.3%, 25% and 4.2% for one, two, three and four years, respectively.
CONCLUSIONS: The rate of AVF creation for end-stage renal disease patients in this setting far exceeds the target goals set forward by the National Kidney Foundation published updated Dialysis Outcomes Quality Initiative (NKF/DOQI) Guidelines and the Centers for Medicaid & Medicare Services Fistula First initiative. This is being achieved with acceptable rates of morbidity and patency. There is room for improvement in postoperative surveillance to increase early detection of failing accesses and allow for increased utility of interventions for assisted patency.