Outcomes After Primary Arterio-venous Fistula Formation for Haemodialysis Access in End Stage Renal Disease: Four Year Follow-up of Patients at a Single Centre in New Zealand. Should we be Prioritising Success Over Vein Conservation in a Population With Short Life Expectancy?

Abstract

Introduction: Despite many studies reporting fistula patency rates and secondary interventions, few studies report longitudinal patient outcomes after first fistula formation. 23% of NZ haemodialysis patients are patients of Counties Manukau Health1. Many of these patients have a short life expectency due to co-morbidities, particularly in the older age group. The concept of conserving vein for future fistulas is therefore less valid. It is desirable to achieve a successful fistula with the first procedure. We assessed whether a vein conserving approach compromised successful outcome in Counties Manukau patients. Methods: A prospective analysis was undertaken of patients whose first fistula was formed between 1st January 2010 and 31st December 2014 by 2 surgeons at Counties Manukau Health, Auckland, NZ. Data collected included; demographics, dialysis status, fistula usage, reasons for not using the fistula, need for further fistula formation and patient outcome as at December 2018. Patients with prior attempt at fistula formation were excluded. The primary outcome was a successful fistula, defined as being the sole method of dialysis for over 6 months. Secondary outcomes were reasons for fistula failure. Results: Overall, 154 patients met inclusion criteria with ages ranging from 20 to 77 years (median 57) with 58% male. 78% were Maori or Pacific, compared with 59% in the NZ haemodialysis population1. Diabetes (71%) was one of the most prevalent co-morbidities, higher than the NZ haemodialysis population (58%). With respect to fistula formation, 2 patients had PTFE grafts, the rest were autologous vein. Radiocephalic fistulas comprised the majority 109(71%), with 17(16%) failed. There were 38(25%) brachiocephalic fistulas formed, with 14(37%) failed. Follow up for 2 patients could not be completed and all patients had a minimum of 4 years follow up. There were 4 outcome categories: 99(64%) fistulas were used until death, transplant or end of follow-up; 17(11%) were used for 6 or more months (median time to failure of 45 months); 32(21%) failed and had further procedures; and 6(4%) never used the fistula. Of the 17 that were used for over 6 months then failed, 1/3 were due to infection and secondary bleeding and 1/3 due to previously unrecognised central stenosis. 6 of the 32 failed fistulas(19%) were ligated due to steal syndrome all of which were brachiocephalic fistulas.16% of all brachiocephalic fistulas. Conclusion: The cohort of 154 patients had 64% requiring only one fistula formation and 11% had successful usage for a median of 45 months. Rates of infection and bleeding were higher than expected. This was surprising given the majority were autologous veins. There were also high rates of undiagnosed central stenosis indicating a need for central venous imaging preoperatively. Brachiocephalic fistulas had16% (6/38) risk of steal syndrome which could be attributed to the high rates of diabetes in the cohort, increasing the risk of microvascular complications. In summary, the current policy of a vein conserving approach is satisfactory. Moving to a larger more proximal vein is associated with increased risk of complication, in particular, steal syndrome. There is a role for further assessment of distal perfusion in these patients. Disclosure: Nothing to disclose