Elements of dialysis nursing practice associated with successful cannulation: result of an international survey



Vascular access (VA) cannulation is an essential skill for dialysis nurses: failure to correctly repeat this operation daily may result in serious complications for the patients. This study investigates if different aspects of arteriovenous fistula and graft cannulation have an effect on the development of acute access complications, which may affect the VA survival.


In April 2009 a cross-sectional survey was conducted in 171 dialysis units located in Europe, the Middle East and Africa to collect details on VA cannulation practices. Information on cannulation retrieved from the survey comprised fistula type and location, cannulation technique, needle size, use of disinfectants and of local anaesthetics, application of arm compression at the time of cannulation, needle and bevel direction, needle rotation, and needle fixation. Five categories of complications were investigated: multiple-cannulation, infiltration, haematoma, haemorrhage and unknown.


There were 10,807 cannulation procedures evaluated in the same number of patients. Of these, 367 showed some kind of complication, the most frequent (33.8%) being the need for multiple-cannulation. The following were associated with a significantly higher odds ratio for occurrence of an acute complication: prescription of back-eye needles, use of rope-ladder cannulation technique, insertion of venous needle as first needle, and rotation of the arterial needle. Use of 16-17-gauge needles was also significantly associated with complications, but this possibly reflects poor quality of the VA.


The risk of an acute VA complication could be reduced with appropriate training of nurses, physicians and patients. This could potentially prolong the VA life.

J Vasc Access 2017; 18(2): 114 - 119





Maria Teresa Parisotto, Francesco Pelliccia, Aileen Grassmann, Daniele Marcelli

Article History


Financial support: No grants or funding have been received for this study.
Conflict of interest: All authors are employees of Fresenius Medical Care.

This article is available as full text PDF.

Download any of the following attachments:


Patients on extracorporeal dialysis treatment depend on well-functioning vascular access (VA) for their survival. The arteriovenous fistula (AVF) presents as the optimal VA, having a lower rate of complications, lower morbidity and fewer maintenance costs compared to central venous catheters (1). In 2006 the National Kidney Foundation Kidney Disease Outcomes Quality Initiative (KDOQI) introduced a list of features that define a well-functioning VA among which was the statement “it can be easily cannulated”, thus shifting focus from the surgical aspects of VA preparation to the way in which the VA is handled in normal clinical practice (1). Although the usual insertion of two needles into the vessels is a basic nursing action, if we consider that it has to be repeated roughly 156 times a year we can understand its importance in the medium-/long-term maintenance of the VA. Cannulation is an essential skill for dialysis nurses: failure to correctly repeat this operation safely and effectively day after day may result in serious complications for the patients. The matter is not only related to expertise: in the current almost “assembly line” factory attitude for dialysis treatment (2), the pressure to execute all operations in the shortest possible time has led to focus more on the time required for an action than on the result thereof. In fact, time of conducting an action can be immediately measured, but the result of the action that is the medium- to long-term damage of the VA cannot be directly related to poor cannulation practices.

AVF or arteriovenous graft (AVG) cannulation, even if usually considered a simple action, is actually a procedure involving several steps. Each step can be executed in a variety of ways. The first is the selection of the needle, which can be with or without back-eye, of bigger or smaller size, and of shorter or longer cannula. Then, after the skin disinfection with different agents, a local anaesthetics can be used. A tourniquet or some form of vessel engorgement technique (e.g., compression of the arm by the patient or a member of the staff) can be applied (or not) to facilitate cannulation, which can be conducted in one of three different ways: area (“one-site-itis”, insertion of the needles in the same general area, session after session), rope-ladder (“site-rotation”, change of the needle placement sites for each dialysis session, choosing sites at a defined distance from the previous one along the vascular access line) or buttonhole (“constant-site”, insertion of the needle in the exact same spot and at the same angle and depth of penetration for consecutive dialysis sessions). Arterial needles can be inserted antegrade (in direction of the blood flow) or retrograde, with the bevel up or down. If the needle is inserted with bevel up, then this needle is often rotated in an attempt to produce a higher blood flow. This rotation might result in additional endothelial trauma (3), increasing the size of the puncture orifice, and so facilitating the exposure to bacterial pathogens and increasing both peri-needle bleeding during treatment as well as bleeding time after needle removal. Finally, needles can be fixed to the patient’s arm in different ways, sometimes codified as Butterfly, Chevron and U-shape and sometimes not.

It is clear that all these different options can affect the success of the cannulation, firstly by avoiding the need to insert additional needles or recannulate after initial failure, but also by avoiding the high-risk complications of haematoma, haemorrhage and blood infiltration.

A few studies already addressed the issue of cannulation practice and associated complications (4-5-6), but convincing evidence for best practices is still lacking. The aim of this study was to investigate if the different aspects (i.e., needle axis rotation, needle fixation, cannulation technique) of AVF and AVG cannulation have an effect on the development of acute access complications. These are defined here as the need for multiple-cannulation, haemorrhage, haematoma, infiltration and unknown.

Patients and methods

In April 2009, a cross-sectional survey was conducted in 171 dialysis units located in Europe, the Middle East and Africa to collect details on VA cannulation practices on a clinic by clinic level. The survey details and results have already been published (7, 8). The association between acute access complications (i.e., multiple-cannulation, haemorrhage, haematoma, infiltration, unknown) observed during the survey and cannulation practice was investigated in all 10,807 patients enrolled in the survey.

Information on cannulation retrieved from the survey comprised fistula type and location, cannulation technique, needle size, use of disinfectants and of local anaesthetic, application of arm compression at the time of cannulation, arterial needle and bevel direction, needle rotation, first needle inserted (arterial/venous), and needle fixation. Five categories of complications were investigated: multiple-cannulation, infiltration (needle dislodgment from inside the fistula/graft causing blood leaks outside of the vessel into the surrounding tissues, swelling and pain in the area), haematoma (results from needle infiltration with signs of bruising to the skin and/or hardened lumps), haemorrhage and unknown.

Data were reported as proportions. A logistic regression model was used to investigate the relationship between variables and the onset of a VA complication on the day of the survey. The following variables were considered: needle size with or without back-eye, skin disinfectant, use of a local anaesthetic, application of tourniquet, cannulation technique (area, rope-ladder and buttonhole), antegrade or retrograde arterial needle insertion with bevel up or down, needle rotation, and needle fixation (i.e., Butterfly, Chevron, U-shape). These needle fixation methods are defined as follows: Butterfly - step 1: apply a strip of tape, sticky side down, around the wings of the needle (Butterfly) and then stick it to the skin; step 2: secure the needle tubing with a strip (covering at least 70% of the tube perimeter before sticking it to the skin. Chevron - step1: tape the top of the wings and fix them to the patients’ skin; step 2: position a second strip under the needle tube with the adhesive facing up, cross the end of the tape to form a Chevron to further secure the wings; step 3: secure the needle tubing with a strip. U-shape - steps 1 and 3 are the same as Chevron; step 2: position a second strip under the needle tube with the adhesive facing up, secure the wings by fixing the tape on top to form a U-shape. Cannulation-related variables that showed an association with the presence of a complication were entered in a multivariable logistic regression model. Results were considered statistically significant for p<.05. SPSS (SPSS Statistics for Windows, Version 21.0, IBM Corp., Armonk, NY, USA) was used for all statistical analyses.


In total, 10,807 cannulation procedures in the same number of patients were observed and included in the current evaluation. A detailed description of the cannulation process by country has been already reported (7). There were 91.2% of cannulations performed on AVFs (of these, 54.1% distal, 45.9% proximal), and 8.8% on AVGs. Table I shows a comparison of cannulation characteristics between AVFs and AVGs. In terms of needle size, the more prevalent dimension was 15-G for AVF and 16-G for AVG. Needles with back-eye were used in 73% (forearm) and 68% (upper arm) of AVF cannulations, and in 35% (forearm) and 47% (upper arm) of AVG cannulations. A local anaesthetic was used in a small number of cases (4.1%-9.4%, depending on VA type and position) without relevant differences between types of VA. The first needle inserted was usually the arterial needle and the bevel direction was up. Arterial needle rotation was performed in about 42% to 50% of the insertions, with the majority of these being inserted in the antegrade direction. The most common cannulation technique applied was area (61%) followed by rope-ladder (28.9%) and buttonhole (6.1%) and the most common needle fixation was “U-shape”.

Comparison in cannulation characteristics in percentages between arteriovenous fistula and arteriovenous graft and between forearm and upper arm

Cannulation practice Arteriovenous fistula Arteriovenous graft
Forearm Upper arm Forearm Upper arm
Needle gauge 14 2.9 2.3 0.0 0.4
15 64.4 62.9 40.0 40.6
16 30.5 32.4 47.7 53.0
17 1.7 1.3 6.2 3.9
Back-eye needle Yes 72.8 67.9 34.8 46.6
No 27.2 32.1 65.2 53.4
Local anaesthetic Yes 8.7 9.4 4.1 6.8
No 91.3 90.6 95.9 93.2
1st needle inserted Arterial 68.3 64.4 53.4 46.2
Venous 31.7 35.6 46.6 58.3
Needle position Bevel up 74.9 73.3 62.6 59.0
Bevel down 25.1 26.7 37.4 41.0
Needle axis rotation Yes 44.3 44.1 50.4 42.3
No 55.7 55.9 49.6 57.7
Direction arterial needle Antegrade 65.5 64.9 54.7 41.0
Retrograde 34.5 35.1 45.4 59.0
Cannulation technique Buttonhole 6.0 7.4 1.6 1.5
Rope-ladder 33.3 25.8 49.2 49.9
Area 60.7 66.8 49.2 48.6
Needle fixation Butterfly 6.0 4.6 1.5 3.4
Chevron 9.8 8.3 10.0 12.3
U-shape 28.8 28.1 24.6 32.6
Others 55.4 58.9 63.8 51.6
Dialysis nurse experience (years) <.5 3.7 3.8 8.4 5.5
0.5-1.0 4.0 4.2 1.5 2.9
1-2 6.3 7.3 5.3 6.8
2-5 20.2 18.7 13.7 20.8
>5 65.8 66.0 71.0 63.9

Out of all observed cannulation procedures, 367 were associated with some kind of complication (multiple-cannulation, haemorrhage, haematoma, infiltration), with the need for multiple-cannulation being the most frequent (33.8%). Figure 1 shows the proportion of cannulation complications by vascular access type and anatomical location: with the exclusion of the rare forearm grafts, no large differences between the various vascular access locations were detected. Table II shows the results of the logistic regression model evaluating the association between cannulation practice and occurrence of an acute complication. Using the needle size of 15-G as reference, the application of 16- or 17-gauge needles was associated with significantly higher odds ratio.

Results of the logistic regression model evaluating nursing practice aspects in respect to the event “cannulation complication”

Nursing practice aspect Odds ratio 95% confidence interval p value
Variable Reference Category Lower limit Higher limit
Needle gauge 15 14 0.545 0.199 1.491 NS
16 1.305 1.016 1.676 0.037
17 4.245 2.548 7.072 <.001
Back-eye needle No Yes 1.879 1.409 2.508 <.001
Cannulation technique Rope-ladder Buttonhole 0.559 0.332 0.942 0.029
Area 0.613 0.482 0.780 <.001
1st needle inserted Arterial Venous 1.677 1.306 2.155 <.001
Needle axis rotation No Yes 1.522 1.206 1.921 <.001
Needle fixation Butterfly Chevron 0.836 0.474 1.475 NS
U-shape 0.754 0.461 1.232 NS
Others 0.561 0.347 0.905 0.018

Proportion of patients with cannulation complications by vascular access type and anatomical location. AV = arteriovenous.

When back-eye needle was used, an 88% significantly higher odds ratio for complication was detected. Using the rope-ladder cannulation technique as reference, the selection of buttonhole and area techniques were associated with a significant 44.1% and 38.7% lower odds ratio, respectively. A significantly higher odds ratio was also estimated when the venous needle was inserted first. In terms of needle fixation, no statistical differences were detected between Butterfly, Chevron and U-shape models, but only for the “other” methods. Disinfectants and application of a compression at the time of cannulation was not significantly associated with the risk of complications. Finally, Figure 2 reports the proportion of patients with different cannulation complications by cannulation technique, with rope-ladder modality more likely to be associated with multiple-cannulation or other complications.

Proportion of patients with cannulation complications by cannulation technique.

Nurse experience with cannulation was as follows: 66% had more than 5 years’ experience; 19% between 2 and 5 years; 7% between 1 and 2 years; and 8% less than 1 year.


In total, 367 procedures out of 10,807 were associated with some kind of complication. To our knowledge, this is the first time that information on prevalence of complications of cannulations has been reported. We believe that this reflects clinical practice because the clinical staff was particularly motivated to record all events in the scope of the study. Although there are no other publications on which to compare our results, we believe that this prevalence of complications is lower than might be expected.

In parallel to our previous study (8) evaluating the impact of cannulation practice on vascular access survival, this study also revealed a significantly higher odds ratio for acute complications when smaller size needles (16- and 17-gauge) were prescribed. The selection of the needle size should depend on the required blood flow (9) but the obvious limitation of the size of the vein has to be considered (1). Therefore, before the application of the tourniquet, the size of the vein should be assessed; the smaller the vein, the smaller should be the size of the needle. The increase of the odd ratio with small needles, is likely an indication of the quality of VA.

To our knowledge, an association between the use of back-eye needles and risk for acute cannulation complications has never previously been reported. It is generally recommended that the arterial needle should have a back-eye in order to prevent suction of the needle to the inner vessel wall (9). In fact, the presence of a back-eye allows an adequate blood flow decreasing the risk of an excessive negative pressure. In addition, it should not require needle rotation, but in our survey 46.8% of the arterial needles with back-eye were rotated, even more than the needles without back-eye (37.5%). In fact, arterial needles without back-eye are mainly used in cannulation with bevel down, not requiring rotation. Therefore, the question of whether or not the higher risk profile found in association with the use of needles with back-eye is of causative nature remains open.

Some authors (10, 11) believe that cannulation of an AVF requires more technical skills than cannulation of an AVG, with reports of higher needle infiltrations in patients with native fistulas. This result was confirmed by van Loon et al (5), who reported a higher percentage of miscannulation as well as cannulation-related complications in patients with AVF compared with AVG. However, in our study the type of VA was not associated with significantly different risk of acute complications. Some cannulation techniques may be of higher complexity. In this study, the area technique showed a significantly lower odds ratio for acute cannulation complications, mainly because of lower risk for unsuccessful cannulations. However, it is important to point out that the results of this study only refer to the occurrence of acute cannulation complications and do not refer to fistula survival. In fact, as reported in a previous publication (8), buttonhole and rope-ladder cannulation techniques were associated with better results than area regarding fistula survival. Van Loon et al (4) already reported a higher frequency of unsuccessful cannulations associated with buttonhole technique, but also less haematoma formation. The rationale could be that unsuccessful cannulation with blunt needles may cause less tissue injury than with the sharp needles used for the rope-ladder and area technique (4). In evaluating the medium- to long-term consequences of different acute cannulation complications, the formation of haematoma may play a major role, thus justifying our previous report of a significantly improved vascular access survival associated with the buttonhole or the rope-ladder compared to the area technique (8). Our current analysis confirmed the significantly lower risk profile of buttonhole cannulation also in terms of acute complications. According to van Loon et al (4), only 8% of patients cannulated with blunt needles ask for local anaesthetic cream compared to 30% exposed to the rope-ladder technique. Surprisingly, and in contrast to other studies (12, 13), this result was not consistent with the evaluated pain score. However, we can speculate that the lower proportion of patients requiring a local anaesthetic may be only partially related to the use of blunt needles, but may also be related to the general lower level of trauma at the needle insertion sites.

When the venous side was cannulated first, a significant 68% higher odds ratio for “cannulation complication” was found. Following this unexpected result, a clarification question was placed to the three countries using this procedure. The use of this procedure was based on the conviction that, since the venous side of the VA is usually more difficult to cannulate, if one succeeds in doing that then one is relatively certain to succeed in cannulating the arterial side, and therefore to perform dialysis. Particularly, in the case of complicated fistulae, the venous side was cannulated first. One can therefore speculate that when the VA was considered problematic (i.e., not functioning well), the cannulation with venous needle first was selected.

Axis needle rotation was associated with a significantly higher odds ratio for acute complications. In detail, rotation practice was associated with a higher frequency of infiltration, haemorrhage and haematoma (1.7% vs. 1.2% without rotation). Therefore, it is likely that the practice of rotation causes an additional trauma to the wall of the vessel, increasing the orifice and allowing the infiltration of blood to the subcutaneous tissue. Session after session, the repetition of this tissue injury may develop into a fibrotic process with consequent earlier termination of the fistula. However, in a previous study, we were unable to detect the impact of needle rotation on fistula survival (8). Nevertheless, on the basis of the results of the current study, we believe that needle rotation should not be recommended.

The procedure to secure needles and blood lines is very important in order to prevent partial or complete needle dislodgment (9), with possible infiltration and formation of haematoma. Our study showed a significantly lower odds ratio for the “other” type of needle fixation compared to Butterfly, with Chevron and U-shape techniques not differing from Butterfly. A second investigation revealed that most of these “other” techniques were actually a modification of the U-shape, which one could approximate as being V-shape. Despite this implication, that fixation with U- or V-shape could be advantageous in avoiding acute cannulation infiltration, concrete conclusions are not possible with our data.

Finally, in relation to nurse experience, we correlated this with the number of complications during cannulation (data not shown) and found that more complications were reported for cannulations performed by experienced nurses. However, as there were no data available linking the access quality with the cannulator, it can only be speculated that the most experienced nurses were dedicated to cannulate the most complicated accesses, hence the higher number of complications reported. In all countries there was a predefined mandatory cannulation training program for all nurses. Some countries have a dedicated vascular access coordinator to manage all issues concerning the vascular access. The cannulation technique used was defined at centre level between the medical director, the head nurse and the vascular access coordinator (where available). The only free decision left in the hands of the individual nurse was the direction of arterial needle (decision influenced by the cannulation space) and the direction of the bevel. Our study has certain limits and strengths. We did not consider patient characteristics and comorbidities that potentially affect the onset of the type of complication considered in our study. Comparison of groups of patients treated with the different cannulation practices in this observational study could therefore be biased. However, since the practice at the time of the survey was still highly related to the single-centre culture and tradition, we believe that this limitation is not major. Another limitation is related to the acute complications referred to as “unknown”, i.e., not referred to as due to multiple-cannulation, haemorrhage, haematoma or infiltration. Since they are not exactly described, we cannot better qualify the severity and importance of this complication group. As strengths of this study we would like to underline the high number of procedures monitored and the large variability in the practice performed in different dialysis units in several countries.

In conclusion, proper cannulation is crucial for the long-term survival of a VA and is a fundamental skill that the nurse must develop. This study identified certain steps in the process of cannulation that could potentially prolong the life of the vascular access. This has consequences not only for patient survival on chronic haemodialysis but also patient quality of life by preventing sufferance due to the creation of additional temporary or permanent vascular accesses.

VA care and management is not without challenges. It requires continuous education of nurses, patients and also physicians. Particularly the first two (nurse and patient) play an important role in ensuring the longevity of the VA. Good knowledge is necessary to enable the nurse to assess, plan, implement and evaluate the care given to patients before, during and after cannulation and to deal with complications properly. However, the impact of education on VA survival and potential reduction of acute cannulation complications has never been properly evaluated. Therefore, more studies should be conducted on this important topic.


We thank the nursing staff from the Fresenius Medical Care dialysis centres participating in this initiative for careful documentation of the clinical practice. The organisational support provided by the country head nurses Joao Fazendeiro (Portugal), Natalie Beddows (UK and Ireland), Malgorzata Liber (Poland), Manuela Moretti (Italy), Francesco Pelliccia (Italy), Emine Unal (Turkey), Mihai Preda (Romania), Asuncion Martinez (Spain), and Sabina Frumen (Slovenia) is gratefully acknowledged. In addition, we specially thank Aleksandra Skinder and Andreja Furlan for their indispensable support.


Financial support: No grants or funding have been received for this study.
Conflict of interest: All authors are employees of Fresenius Medical Care.
  • 1. Vascular Access Work G; Vascular Access 2006 Work Group. Clinical practice guidelines for vascular access. Am J Kidney Dis 2006 48 Suppl 1 S176 S247 Google Scholar
  • 2. Wilson B Harwood L Oudshoorn A Moving beyond the “perpetual novice”: understanding the experiences of novice hemodialysis nurses and cannulation of the arteriovenous fistula. CANNT J 2013 23 1 11 18 Google Scholar
  • 3. Ball L Dilon T Dinwiddie L Holland J Vascular access. Core Curriculum for the Dialysis Technician - 5th Edition: A comprehensive review of hemodialysis. Madison: Medical Education Institute 2012 p. 121 160 Google Scholar
  • 4. van Loon MM Goovaerts T Kessels AG van der Sande FM Tordoir JH Buttonhole needling of hemodialysis arteriovenous fistulae results in less complications and interventions compared to the rope-ladder technique. Nephrol Dial Transplant 2010 25 1 225 230 Google Scholar
  • 5. van Loon MM Kessels AG Van der Sande FM Tordoir JH Cannulation and vascular access-related complications in hemodialysis: factors determining successful cannulation. Hemodial Int 2009 13 4 498 504 Google Scholar
  • 6. van Loon MM Kessels AG van der Sande FM Tordoir JH Cannulation practice patterns in haemodialysis vascular access: predictors for unsuccessful cannulation. J Ren Care 2009 35 2 82 89 Google Scholar
  • 7. Gauly A Parisotto MT Skinder A et al. Vascular access cannulation in hemodialysis patients - a survey of current practice and its relation to dialysis dose. J Vasc Access 2011 12 4 358 364 Google Scholar
  • 8. Parisotto MT Schoder VU Miriunis C et al. Cannulation technique influences arteriovenous fistula and graft survival. Kidney Int 2014 86 4 790 797 Google Scholar
  • 9. van Loon M How to improve vascular access care. Contrib Nephrol 2015 184 222 233 Google Scholar
  • 10. Allon M Robbin ML Increasing arteriovenous fistulas in hemodialysis patients: problems and solutions. Kidney Int 2002 62 4 1109 1124 Google Scholar
  • 11. Lee T Barker J Allon M Needle infiltration of arteriovenous fistulae in hemodialysis: risk factors and consequences. Am J Kidney Dis 2006 47 6 1020 1026 Google Scholar
  • 12. Ball LK The buttonhole technique for arteriovenous fistula cannulation. Nephrol Nurs J 2006 33 3 299 304 Google Scholar
  • 13. Verhallen AM Kooistra MP van Jaarsveld BC Cannulating in haemodialysis: rope-ladder or buttonhole technique? Nephrol Dial Transplant 2007 22 9 2601 2604 Google Scholar



  • Fresenius Medical Care, Bad Homburg - Germany

Article usage statistics

The blue line displays unique views in the time frame indicated.
The yellow line displays unique downloads.
Views and downloads are counted only once per session.

No supplementary material is available for this article.