Is colour duplex ultrasound more sensitive than clinical assessment in predicting early failure of primary arteriovenous fistulae (AVF)?

INTERVENTION: Purpose of Study: This study is intended to improve the service available to patients attending hospital for chronic kidney disease (CKD). Many patients suffering from CKD will require some sort of kidney replacement therapy. Some may be fortunate enough to receive kidney transplant, however those that are not so fortunate will need dialysis of some kind. Dialysis is available in two ways: Peritoneal and Haemodialysis (HD). This study is concentrated on those preparing for HD. In order to begin dialysis a 'man‐made' short circuit of the circulatory system must be created. This involves the connection of a vein to the side of an artery in the arm of the patient. This connection allows the development of the vein into a significantly larger vessel that allows easy needling during dialysis. This also means that the volume of flow through this connection, known as an arteriovenous fistula (AVF), also increases dramatically. It is this high blood flow that allows the patient to receive dialysis in just a few hours, when the flow of blood is reduced the time taken to dialyse is vastly increased. The development of these AVF is not always simple and they can often fail, sometimes only minutes after surgery. In order to improve the chances of survival we would like to investigate the protocols currently used to survey these AVF during development and assess the use of a new protocol and technique. Currently clinical assessment of the developing AVF occurs around two weeks post operation. Any problems identified at this stage mean the patient can be referred for an alternative investigation. This alternative investigation is duplex ultrasound. Evidence supports ultrasounds ability to identify accurately any abnormality in the developing AVF without the need for an invasive technique. This study aims to compare the ability of clinical assessment and ultrasound to accurately predict the outcome of a developing AVF. Whilst doing this we intend to identify the most accurate ti CONDITION: Cardiovascular: Primary arteriovenous fistula ; Circulatory System ; Primary arteriovenous fistula PRIMARY OUTCOME: 1. The specificity and sensitivity of both clinical assessment and ultrasound in predicting AVF success or failure; 2. ROC curves will be used to identify the most accurate ultrasound criteria found for predicting AVF failure or success; 3. ROC curves will also be used for identifying the optimum time for ultrasound surveillance. SECONDARY OUTCOME: Not provided at time of registration INCLUSION CRITERIA: Both groups will be recruited in the same way through the renal consultants and will be randomly allocated to one of two groups. It is predicted that the sample size will be no more than 144 patients in the first year as this is the predicted maximum number of patients requiring primary AVF in this timescale. It is predicted that the final number of patients required for the sample will be approximately 50 per group as a result of a retrospective study performed previously by Robbin et al (2002). INCLUSION CRITERIA: 1. Adult patient requiring primary AVF 2. No previous AVF or dialysis via central line 3. Upper limb vein/artery suitable for AVF formation 4. The patients' eGFRml/min/1.73m^2 must be greater than 15 to minimise likelihood of access being required before maturation of primary AVF