BACKGROUND: The patients diagnosed with end stage renal disease require dialysis and for that they need to have a vascular access for hemodialysis (HD) placed. Vascular access complications are the most common cause of hospitalization among patients in HD. An AV fistula (AVF) is a surgical connection made between an artery and a vein, created by a vascular specialist, typically placed in an arm. AV fistulas are the preferred vascular access for long-term dialysis. In order to be able to carry out an adequate and uncomplicated dialysis treatment, two needles have to be placed in the fistula. It requires specialized technical ability to install well-functioning needles in the vessels of an AVF. The buttonhole technique is one of the two recommended techniques. For the buttonhole technique, two puncture sites are selected in the fistula. Here, exactly in the same spot a needle is inserted with the same angle and direction until a fibrous tunnel is formed, like a hole for an ear ring. Cannulation is now possible with blunt needles which are gentler and reduce complications. The time required to create a buttonhole tunnel is 6-12 cannulations, ie 6-12 dialyses. It is crucial for the future survival of the tunnel tracks, that a maximum of one to two persons cannulates until the track is created. This is a logistic challenge in a busy dialysis unit and may result in using a cannulation technique that is not recommended. A less time consuming method to create the buttonhole tunnel track may increase the use of the technique. A new method has been used in several dialysis units in Denmark. Using this method the tunnel tracks are created in 1-3 dialyses by repeated cannulations (4-6 needles one at a time) in the same two puncture sites in the fistula. The experiences so far indicates that this method reduces the dialysis sessions needed to create the tunnel tracks with 4-10 sessions. Thus, the logistic challenges of ensuring continuity in persons creating the tunnel track will be reduced. The purpose of the research project is to investigate whether a new method for creating buttonhole tunnels will:

• Increase the number of well-functioning buttonholes.
• Be less painful for the patient.
• Reduce the number of dialysis needed to create the buttonhole tunnel track.
• Cause unchanged or fewer fistula associated complications and infections.

Background: See above. Aims:To study whether a new method of creating a buttonhole tunnel tracks can reduce the incidence of buttonhole technique failure and the patients pain with needling.

Patients and Methods: A randomized, multi-center study enrolling 42 patients in hemodialysis from two HD units in Denmark. The study will begin end of May 2019.

Participants are randomized into 2 groups where cannulation will be carried out by a small group of dialysis nurses with experience and skills in buttonhole cannulation. In both groups the same nurse will carry out the cannulation the first four dialysis sessions. At the following sessions the cannulation will be carried out by the nurse/nurses responsible for the patient. The procedures throughout the study will be the same in both participating centers.

The control group (K): creating buttonhole tunnels according to the standard method, ie. cannulation with a single sharp cannula in the AVF at each dialysis for 6-12 dialyses until a blunt needle can be inserted.

The intervention group (I): creating buttonhole tunnels according to the new method, ie. repeated cannulation with several sharp needles (4-6 pieces) at each dialysis for 1-3 dialyses until a blunt cannula can be inserted.

The creation of a buttonhole tunnel track is defined as completed when two consecutive dialyses have been carried out with no use of sharp needles, however no longer than 24 dialysis sessions.Thus, the period of time will be of different length for each participant. In the period until the tunnel is created, data will be collected at each dialyses regarding number of cannulations, type of needle (sharp or blunt), cannulator and patient-experienced pain with needling.

The primary outcome is buttonhole technique failure defined as the need for cannulations carried out using sharp needles in the buttonhole for three consecutive dialyses or in case lack of ability to insert blunt needles in the buttonhole for 2 weeks.

Follow-up from the date of the completed creation of the buttonhole tunnel track in both groups:

12 month ongoing follow-up of the primary outcome buttonhole survival and the secondary outcomes AVF associated infections, complications and AVF survival.

Sample size:The power calculation is based on the expected incidence of the primary outcome measure: buttonhole technique failure. Based on calculations in the two dialysis centers from randomly selected patients with buttonhole tunnels made using the intervention and the control method, it appears that the estimated buttonhole technique failure is 10 % and 30 %, respectively.

Drop-off rate is estimated to 30%. The number of participants for the study is calculated at 21 subjects in each group to achieve a power of 0.8 with a type error rate of 0.5.

Intention-to-treat analysis will be carried out. Censoring in case of death, if the person wants to withdraw from the study or moves, AV fistula failure for non-buttonhole associated causes and kidney transplant.

Statistical analysis:The baseline characteristics of the study population will be presented as counts and proportions and compared between randomization groups with chi-square test / Fisher's exact test (depending on number of observations) for categorical variables. Continous variables will be presented as mean (SD), respectively, median (IQR) depending on normality of data and compared between groups by t-test, respectively, Wilcoxon rank-sum test. Normality will be evaluated by quantile plots and Shapiro-Wilks test. Moreover, repeated measurements will be compared by mixed effects models. Time to event outcomes will be compared by the Kaplan-Meier method and tested by log-rank test, furthermore discrete time survival methods will be applied as a sensitivity analysis. P-values <0.05 are regarded as statistically significant. Follow-up will be censored at 1 year. Statistical analysis will be performed using Stata, version 11.0 (StataCorp, College Station, Tex., USA).

Inclusion Criteria:

• chronic hemodialysis patients
• at least 5 hemodialysis treatments with a blood flow ≥ 250 ml/min have been carried out
• ability to speak and understand danish

Exclusion Criteria:

• patients with an artificial heart valve
• patients in high dose immunosuppressive treatment
• patients with infections in their buttonholes
• patients who are unable to provide informed consent