The requirements for haemodialysis are increasing. There is now acceptance that the most durable AV fistulae are those created from the patients own veins. Use of the radiocephalic arteriovenous fistula as an autologous vascular access dates back to the 1960's. (Brescia MJ, Cimino JE. Appel K, Hurwich BJ. Chronic hemodialysis using venipuncture and a surgically created arteriovenous fistula. N Engl J Med 1966; 275: 1089 - 1092.) Once established, it has good long term survival and a low complication rate. However, the success rate at creation of a useable AV fistula are not good. In a meta-analysis published in 2005, the primary failure rate of a radiocephalic fistula was 15.3% and the primary and secondary patency rates were 62.5% and 66.0% at one year. (Rooijens PPGM, Tordoir JHM, Stijnen T, Burgmans JPJ, Smet AAEA and Yo TI. Radiocephalic wrist arteriovenous fistula for hemodialysis: meta-analysis indicates a high primary failure rate. Eur J Vasc Endovasc Surg 2004; 28, 583-589). Reasons for this are multiple but it appears as though the main determinants are the quality of the patients veins and the technical skill of the surgeon. Low patency rates are also seen in females and those with small veins and arteries. Creating an anastomosis between a small vein and artery is technically challenging. This is traditionally preformed with a continuous prolene suture. Recently, Medtronic have been marketing the U Clip Anastomotic Device. This is a self closing devise which could be called a vascular clip. Reported advantages of this clip include better patency rates and faster operations. However, to our knowledge there is no randomised evidence to support this with regards to AV fistulas.

The vascular anastomosis originally developed by Alexis Carrel used interrupted sutures. Over time this was replaced by a continuous suture technique which was felt to be quicker and had similar patency rates. However, microvascular, paediatric and neurovascular surgeons have shown that there are advantages to an interrupted technique. (Hattori H, Killen DA, Green JW. Influence of suture material and technic on patency of anastomosed arteries of less than 1.5mm. Am Surg. 1970; 36: 352 - 354. Cobbett JR. Microvascular surgery. Surg Clin N Am. 1962; 47: 521.). The reasons for the improved patency with interrupted anastomoses are firstly increased anastomotic compliance and flow rate and secondly elimination of the pursestring effect and puckering seen with continuous sutures.

In a prospective but non randomised study to compare interrupted U clips with historical published results, the patency rate of coronary anastomoses was 100% at 6 months using the U clips compare to a patency rate of 90 -100% in the published series. (Wolf RK, Alderman EL, Caskey MP et al. Clinical and six month angiographic evaluation of coronary arterial graft interrupted anastomoses by use of a self closing clip device: a multicentre prospective clinical trial. J Thorac cardiovasc Surg 2003; 126(1): 168 - 178.)

Reports from single centres have concluded that the U Clips offer the opportunity to create superior interrupted anastomoses for AV fistula, even in patients who would otherwise be considered poor candidates for fistula creation. (Ross JR. Creation of native arteriovenous fistulas with interrupted anastomoses using a self closing clip device - one clinics experience. Journal of vascular Access 2002; 3: 140 - 146). In this report with small numbers, the radiocephalic fistulas had an 8 week maturation rate of 93%. Of the 28 patients having a radiocephalic fistula, 10 had veins of between 1.0 - 1,5mm diameter.

The aim of this pilot study will be to assess if there is a clinical difference in the maturation rates of autologous AV fistulae when a clipped anastomosis is compared to a conventional prolene anastomosis.

Method. A prospective, randomised, blinded study to compare continuous prolene anastomosis with interrupted U clip anastomosis in creation of autologous AV fistula.

Recruitement. Patients will be referred by the nephrologists to the vascular surgeons for assessment prior to creation of an AV fistula in the usual way. Current unit guidelines are that all patients have a pre operative duplex scan to assess the superficial veins. If the patient meets the entry criteria and there are no exclusion criteria, the patient will be consented to be involved in the trial.

The patient will be admitted for surgery and anaethetised in the usual way. Once the vein and artery have been dissected and prepared for anastomosis a sealed envelop will be opened. This will instruct the surgeon to perform the anastomosis with either a conventional continuous prolene suture or interrupted U Clips. The envelop will also contain a study number which is recorded on the operation note. The operation not must no include the words prolene anastomoiss or clipped anastomosis. In this way subsequent assessment by nephrologists will be blinded.

Patients will then undergo follow up in the usual way by their nephrologists. The nephrologist should then report the subsequent outcome of the fistula and patients will undergo clinical follow up alone.

Randomisation. This will be performed using an electronic random number generator. This will result in a study number and surgeon instruction being placed in an opaque sealed envelop which will only be opened in theatre just before the anastomosis is performed. It will be a 1:1 randomisation.

Primary outcome measure. A functioning fistula which is used for haemodialysis on three or more occasions.

Secondary outcome measures.

1. Fistula patency as assessed clinically.
2. Need for fistula intervention in order to maintain patency.

Inclusion Criteria:

1. Requirement for AV fistula creation for haemodialysis
2. Pre operative upper limb venous and arterial duplex scan showing intended arterial lumen diameter of 2mm or more and vein diameter of 3mm or more.

Exclusion Criteria:

1. Use of prosthetic graft for fistula creation.
2. Donor arterial lumen diameter less then 2mm
3. Recipient venous diameter less than 3mm.
4. Proximal venous thrombosis.