clinical practice guideline on vascular access for ... · web viewthe point estimate favoured no...

Clinical Practice Guideline on Vascular Access for Haemodialysis in Adults - Part 1

Peri- and postoperative aspects of arteriovenous fistulas and grafts

of 226

Table of Contents

1. Abbreviations and Acronyms 3

2. Definitions 3

3. Preface 5

4. Summary of the recommendations 6

5. Composition of the Guideline Development Group 9

6. Conflict of interest policy and declaration of interest forms 12

7. Purpose and Scope of this guideline 13

8. Methods for guideline development 15

9. Chapter 1. Medical treatments for promoting AV fistula maturation 20

10. Chapter 2. Surgical and endovascular interventions for promoting AV fistula maturation 26

11. Chapter 3. Surgical and endovascular interventions for non-maturing AV fistulas 29

12. Chapter 4. Self-administered interventions for AV fistula maturation 32

13. Chapter 5. Perioperative prophylactic antibiotics for preventing AV access infection 35

14. Chapter 6. Timing of first cannulation 37

15. Chapter 7. Vascular access surveillance 42

16. Chapter 8. Medical treatments for maintaining long term AV access patency 49

17. Chapter 9. Cannulation techniques for AV fistulas 54

18. Chapter 10. Needle types for AV fistulas 59

19. Chapter 11. Timing of intervention for AV fistula thrombosis 61

20. Chapter 12. Surgical and endovascular interventions for AV access thrombosis 64

21. Tables 67

22. Figures 73

23. Acknowledgements 76

Supplement 1| Guideline development group area of expertise 77

Supplement 2| Declaration of interest statements 83

Supplement 3| Review Questions - PICOM Format 103

Supplement 4| Search Strategies 111

Supplement 5| Study selection flow diagrams 124

Supplement 6| Summary evidence tables 132

Supplement 7| Internal Review 205

Supplement 8| External Review 207

24. References 220

of 226

1. Abbreviations and Acronyms

AV ArteriovenousCPG Clinical Practice GuidelineCKD Chronic Kidney DiseaseCSN Canadian Society of NephrologyDOPPS Dialysis Outcomes and Practice Patterns StudyEBPG European Best Practice GuidelinesEDTNA/ERCAEuropean Dialysis and Transplant Nurses Association/European Renal Care

AssociationERBP European Renal Best PracticeESKD End-stage kidney diseaseESVS European Society for Vascular SurgeryGEMAV Grupo Español Multidisciplinar del Acceso VascularGRADE Grading of Recommendations Assessment, Development and EvaluationHR Hazard RatioI² I squareIQR Interquartile rangeIRR Incidence Rate RatioKDIGO Kidney Diseases Improving Global OutcomesNKF-KDOQI National Kidney Foundation – Kidney Diseases Outcomes Quality InitiativeKHA-CARI Kidney Health Australia – Caring for Australasians with Renal InsufficiencyMD Mean DifferenceN Number ofNICE National Institute of Clinical ExcellenceOR Odds RatioP P-valuePTFE PolytetrafluoroethyleneRCT Randomised Controlled TrialRR Relative Risk95% CI 95% Confidence IntervalSIGN Scottish Intercollegiate Guidelines NetworkSD Standard DeviationSMD Standardised Mean DifferenceUKRA UK Renal AssociationVAS Vascular Access Society

2. DefinitionsInterpreting evidence in the AV access literature is challenged by the heterogeneity in terminology and the lack of standardisation in outcomes. Below are listed some of the terms used in this guideline and how they have been interpreted in the context of this document. Because guideline development necessarily relies on aggregate data from systematic reviews and other individual studies, we can only hope to provide the user with conceptual definitions for certain concepts and outcome domains. At present, there is insufficient consensus to go beyond that point and define specific outcome measures or measurements.

AV access overarching term referring to both AV fistulas and AV grafts

AV access thrombosis blood clot obstructing the AV access; indicates loss of anatomic, hemodynamic, and clinical patency

of 226

AV fistula surgically created autogenous vascular access used for chronic haemodialysis consisting of an anastomosis between an artery and a vein, with the vein serving as the accessible conduitSynonym: native AV fistula

AV graft surgically created vascular access used for chronic haemodialysis whereby an artificial or biological prosthetic segment is used to connect an artery and vein, with the prosthetic segment serving as the accessible conduit

Cannulation placement of a dialysis needle in the AV access to provide haemodialysis

Clinical monitoring clinical assessment of an AV access at regular intervals; it is distinct from technical surveillance; it includes examination of the access AV thrill and bruit, haemostasis time after needle removal and outflow appraisal after arm elevation.

Clinical practice guideline a set of statements that include recommendations intended to optimize patient care that are informed by a systematic review of evidence and an assessment of the benefits and harms of alternative care optionsSynonym: guideline

Maturation process leading to a newly created AV access being usable for haemodialysis; it encompasses enlargement and thickening of the draining fistula vein, increases in the blood flow, and absence of thrombosis and bleeding as mechanisms of AV access failure. Synonym: suitability for dialysis.

Recommendations graded statements within a clinical practice guideline intended to optimize patient care that are informed by a systematic review of evidence and an assessment of the benefits and harms of alternative care options

Patency see primary unassisted patency or secondary patency

Pre-emptive intervention intervention aiming to resolve a stenosis or other problem in an AV access that is still adequately providing dialysis; has the intention to avoid the AV access becoming dysfunctional

Primary AV access failure an AV access that, despite radiological or surgical intervention, cannot be used successfully for dialysis by a given time point (usually up to three months) following its creationSynonym: dialysis suitability failure

Primary unassisted patency the time of AV access creation or placement until any first intervention (endovascular or surgical) to maintain or restore blood flow, or first occurrence to AV access thrombosis.Synonym: intervention-free AV access survival; primary patency

Secondary patency the time of AV access creation or placement until AV access abandonment or permanent loss of the AV access. Synonym: cumulative AV access survival; secondary assisted patency

of 226

Surveillance overarching term referring to both clinical monitoring and technical surveillance of an AV access; it includes haemodialysis parameters such as pump speed, dialyser inlet and transmembrane pressure, and indices of dialysis adequacy (Kt/V urea); sequential measurements with trend analysis of intra-access flow, dynamic or static dialyser outlet pressure, AV access recirculation, or AV access duplex ultrasound assessment.

Technical surveillance assessment of an AV access at regular intervals using specialised apparatus; distinct from clinical monitoring

3. PrefaceVascular access remains one of the most challenging aspects of renal replacement therapy. If the vascular access does not function properly, haemodialysis will not remove uraemic retention solutes adequately, and if all access possibilities have been exhausted, haemodialysis is no longer possible. The consequences for patient survival may be severe, if kidney transplantation or peritoneal dialysis are not options either. Every professional who has been active in haemodialysis long enough will remember at least a few patients in whom haemodialysis had to be abandoned for lack of access. They will also remember many more in whom a series of procedures was necessary, often without ensuring adequate dialysis for prolonged periods. For patients, the ‘umbilical cord’ keeping them alive, can be a constant source of stressful experiences; [1].In 2007, during the second round of recommendations for haemodialysis from the European Best Practice Guidelines (EBPG) – the predecessor of the current European Renal Best Practice (ERBP) – the first set of essentially clinically oriented vascular access recommendations was drafted by a small group of experts in vascular access surgery, interventional radiology and haemodialysis [2]. Guideline development has changed profoundly since then, with more rigorous methodology having been introduced and a greater emphasis put on evidence-based medicine [3].One of the caveats in the present clinical practice guideline (CPG) is that even today high quality evidence-based data on vascular access are scarce, partly because there are still too few sufficiently powered and well-designed controlled trials, and partly because adoption of evidence-based medicine in the field of vascular access is maturing and changing the landscape. It is well recognised that heterogeneity of the patient samples studied and the many associated confounders may bias study results, including: differences in surgical procedures, in skills and experience; differences in patient education; variability in patient genetic predisposition of thrombogenicity; variation in cannulation procedures, uraemic status, and vessel quality; and many others. It made offering strong treatment guidance difficult. However, in addition to helping clinicians make decisions based on what is known today, we hope this text will stimulate researchers to explore what is still unknown, and the nephrological community at large to develop uniform definitions and assess more clinically relevant vascular access outcomes [4].This text is intended for nephrologists, and also for the other stakeholders in the field whose participation was sought during the development process: dialysis nurses, vascular access surgeons, radiologists, researchers, pharmacists, and importantly, patients and their carers [5]. It specifically covers peri- and postoperative aspects of arteriovenous (AV) fistulas and grafts. A second part – under development when this guideline went to press – will cover aspects related to access choice, preoperative vessel assessment and central venous catheters. We hope the current and planned CPG will help assist the professional community in making decisions about vascular access processes, pathways and care; help patients and carers gain insight; and facilitate joint decision-making in this field.

of 226

4. Summary of the recommendationsChapter 1. Medical treatments for promoting AV fistula maturation1.1. We suggest any decision to give aspirin, ticlopidine, or clopidogrel in adults with end-stage kidney

disease during the first two months after AV fistula creation for the sole purpose of improving maturation, must balance a reduction in thrombosis against uncertain effects on maturation and bleeding. (2C)

1.2. We suggest any decision to give perioperative heparin in adults with end-stage kidney disease during AV fistula creation, must balance an increase AV fistula patency at one month against an important increase in bleeding complications. (2C)

1.3. We suggest any decision to apply far infrared therapy in adults with end-stage kidney disease during the first three months after AV fistula creation, must balance a possible reduction in thrombosis against uncertain effects on maturation and bleeding. (2C)

1.4. There are insufficient RCT data to make a recommendation for ticagrelor, prasugrel, dipyridamole, sulphinpyrazone, warfarin or other oral anticoagulants, fish oil, statins, vonapanitase, glycerine trinitrate, iontophorectic injection of Salvia miltiorrhiza, or prednisolone for improving AV fistula maturation in adults with end-stage kidney disease. (-D)

Advice for clinical practice Do not stop mono-antiplatelet treatment in adults undergoing AV access creation.

Chapter 2. Surgical and endovascular interventions for promoting AV fistula maturation2.1. We suggest using regional block anaesthesia rather than local anaesthesia for AV fistula creation

in adults with end-stage kidney disease. (2C)2.2. We suggest there is insufficient evidence to support end-of-vein to side-of-artery over side-of-vein

to side-of-artery anastomosis for AV fistula creation in adults with end-stage kidney disease. (2C)

Chapter 3. Surgical and endovascular interventions for non-maturing AV fistulas 3.1. We suggest there is insufficient evidence to support open surgical over endovascular interventions

as the preferred treatment for non-maturing AV fistulas in adults with end-stage kidney disease. (2D)

Advice for clinical practice Decisions on how to treat non-maturing AV fistulas are likely best based on local resources,

experience, and success rates. Institutions likely benefit from building a dedicated multidisciplinary vascular access team, with

clinical experience in various techniques available for non-maturing AV fistulas.

Chapter 4. Self-administered interventions for AV fistula maturation4.1. We suggest a standardised exercise program involving hand and arm exercises may improve AV

fistula maturation in adults with end-stage kidney disease. (2C)4.2. There is insufficient evidence to support specific exercise programmes or physical interventions to

promote AV fistula maturation in adults with end-stage kidney disease. (-D)

Advice for clinical practice Involving patients more actively in preparing for haemodialysis may improve self-management

skills, health literacy and thereby well-being.

of 226

Chapter 5. Perioperative prophylactic antibiotics for preventing AV access infection 5.1. We recommend giving preoperative antibiotic prophylaxis for AV graft insertion in adults with

end-stage kidney disease. (1C) 5.2. We suggest giving preoperative antibiotic prophylaxis for complex AV access procedures in

adults with end-stage kidney disease. (2D) 5.3. We suggest not giving preoperative antibiotic prophylaxis for simple AV access procedures in

adults with end-stage kidney disease. (2D)

Advice for clinical practice Simple AV access procedures include the creation of a native radio-cephalic or native brachio-

cephalic AV fistula. Complex AV access procedures include those that are not considered simple.

Chapter 6. Timing of first cannulationAV fistulas6.1. In adults requiring haemodialysis, we suggest AV fistulas can be cannulated four weeks after

creation if they are considered suitable for cannulation on clinical examination. (2C)6.2. In adults requiring haemodialysis, we recommend against cannulating AV fistulas sooner than two

weeks after their creation. (1B)6.3. In adults requiring haemodialysis, we suggest against cannulating AV fistulas in between two and

four weeks after their creation, unless this can avoid placement of a central venous catheter for haemodialysis. (2C)

AV grafts6.4. In adults requiring haemodialysis, we recommend ‘early cannulation type’ AV grafts can be

cannulated as soon as wound healing permits. (1B)6.5. In adults requiring haemodialysis, we suggest against cannulating a ‘standard type’ AV graft

sooner than two weeks after insertion, unless this can avoid placement of a central venous catheter for haemodialysis. (2B)

Advice for clinical practice In practice, suitability for cannulation on clinical examination is determined by the presence of a

palpable vein and good thrill. If clinical examination is inconclusive, ultrasound with flow measurement may help in deciding

whether or not to cannulate. Bedside ultrasound guided cannulation may be helpful in avoiding complications and decreasing

the number of failed cannulations. Using single needle dialysis, low dialysis blood flows and smaller needles (17 gauge) may prevent

harm to AV fistulas, which are cannulated early. Wound healing refers to the tissue around the body of the graft, rather than the incision site.

Chapter 7. Vascular access surveillanceAV fistulas7.1. We suggest the evidence for technical surveillance in addition to clinical monitoring of a

functional AV fistula to detect and pre-emptively correct a haemodynamically important AV access stenosis in adults is inconclusive and needs more research. (2C)

AV grafts7.2. We suggest against technical surveillance in addition to clinical monitoring of a functional AV

graft to detect and pre-emptively correct a haemodynamically important AV access stenosis in adults, unless it occurs in the context of a clinical study. (2C)

Chapter 8. Medical treatments for maintaining long term AV access patencyAV fistulas8.1. We suggest any decision to give fish oil to adults with end-stage kidney disease in the year after

of 226

AV fistula creation, must balance improved patency at one year against an unknown risk of bleeding and other side effects. (2C)

8.2. We suggest far infrared therapy may be considered for improving long term AV fistula patency in adults with end-stage kidney disease. (2C)

8.3. There are insufficient RCT data to make a recommendation for aspirin, clopidogrel, ticlopidine, warfarin, sulphinpyrazone, vonapanitase, beraprost sodium, cholecalciferol, statins, dipyridamole or dipyridamole combined with aspirin to be given for maintaining long-term AV fistula patency in adults with end-stage kidney disease. (-D)

AV grafts8.4. We recommend against warfarin in combination with antiplatelet agents, and against clopidogrel

in combination with high dose aspirin for reducing AV graft thrombosis in adults with end-stage kidney disease. (1C)

8.5. We suggest any decision to give fish oil in the year following AV graft creation in adults with end-stage kidney disease must balance any improvement in graft patency at one year against an unknown risk of bleeding. (2C)

8.6. There are insufficient RCT data to make a recommendation for aspirin, clopidogrel, ticlopidine, warfarin, beraprost sodium, statins, dipyridamole or dipyridamole combined with aspirin to be given for maintaining long-term AV graft patency in adults with end-stage kidney disease. (-D)

Chapter 9. Cannulation techniques for AV fistulas9.1. We suggest against using area technique for cannulating AV fistulas in adults treated with

haemodialysis. (2D) 9.2. We suggest using either a rope ladder or buttonhole technique for cannulating AV fistulas in

adults treated with haemodialysis, and letting the choice be dependent on local expertise and AV fistula characteristics. (2D)

Advice for clinical practice Antiseptic measures and practical aspects of the cannulation procedure are important in reducing

the infection risk associated with buttonhole cannulation. AV grafts are usually only cannulated using a rope ladder technique.

Chapter 10. Needle types for AV fistulas 10.1. We suggest using either sharp needles or plastic cannulas for cannulating AV fistulas in adults

treated with haemodialysis. (2C)10.2. We recommend using blunt needles only for buttonhole cannulation of AV fistulas in adults

treated with haemodialysis. (1D)

Advice for clinical practice A quality improvement program including recording and monitoring of the needle types and

cannulation techniques alongside with AV access outcomes can help monitor quality, guide changes in cannulation practice if needed, and improve quality of vascular access care.

AV grafts are usually only cannulated using sharp steel needles.

Chapter 11. Timing of intervention for AV fistula thrombosis11.1. We suggest attempting to declot a thrombosed AV fistula in adults as soon as possible under

optimal conditions and before the next haemodialysis treatment. (2D)11.2. We suggest attempting to declot a thrombosed AV fistula in adults, even if there has been a

delay of days to weeks. (2D)

Chapter 12. Surgical and endovascular interventions for AV access thrombosis12.1. We suggest the choice between surgical and endovascular interventions for AV access

thrombosis be defined by the condition of the patient and their vascular access, as well as local expertise, as there is no evidence one approach improves outcomes more than any other. (2B)

of 226

5. Composition of the Guideline Development GroupERBP’s Advisory Board appointed the co-chairs and invited a small group of content experts to a steering committee to direct the guidelines development process. These content experts were chosen based on their previous involvement with the EBPG guideline, or their close association with national or international vascular access societies [3]. The group was supplemented with selected members of ERBP’s methods support team to supervise the project and provide methodological expertise in guideline development throughout the process. The steering committee convened in May 2013 and February 2014 and decided on the composition of the Guideline Development Group, taking into account the clinical and research expertise of each proposed candidate, and their willingness to invest the necessary time and effort to perform the task according to the proposed deadlines and the agreed methodology. The group ultimately consisted of 44 participants, including 25 nephrologists, nine surgeons, three radiologists, five researchers, two nurses and eight methodologists (categories not mutually exclusive). It included 29 men, and 15 women (Supplement 1 Guideline development group area of expertise).

Guideline Development Group

Julien Al Shakarchi Registrar Vascular Surgery, ReDVA fellow, West Midlands Deanery, Birmingham, United Kingdom.

Paul BergerConsultant Vascular Surgeon, Zilveren Kruis, Leiden, The Netherlands.

Deirdre CassidyTechnology Leader, GE Healthcare, Chalfont St. Giles, United Kingdom.

Tze Yuan ChanConsultant Interventional Radiologist, Royal Liverpool University Hospital, United Kingdom.

Annemieke DhondtConsultant Nephrologist, Ghent University Hospital, Ghent, Belgium.

Tevfik EcderNephrologist, Istanbul Bilim University School of Medicine, Division of Nephrology, Istanbul, Turkey.

Pietro FinocchiaroConsultant Nephrologist, Nephrology, Dialysis and Transplantation Unit, G.O.M., Reggio Calabria, Italy.

Maurizio GallieniDirector, Nephrology and Dialysis Unit, ASST Fatebenefratelli Sacco, Milano, Italy. Associate Professor in Nephrology, Department of Clinical and Biomedical Sciences “Luigi Sacco”, University of Milano, Italy.

Jennifer HankoConsultant Nephrologist at Belfast Health and Social Care Trust, Belfast, United Kingdom.

Sam HeyeInterventional (Neuro)radiologist, department of Radiology, Jessa Hospital, Hasselt, Belgium.

of 226

Markus Hollenbeck Head of Renal Unit and KfH Kidney centre, Knappschaftskrankenhaus Bottrop, Bottrop, Germany.

Jose IbeasConsultant Nephrologist. Department of Nephrology, Parc Taulí Hospital Universitari, Institut d’Investigació i Innovació Parc Taulí I3PT, Universitat Autònoma de Barcelona, Sabadell, Barcelona, Spain.President-elect of the Vascular Access Society.

Nicholas InstonClinical Lead and Consultant Renal Transplant and Vascular Access Surgeon, University Hospital Birmingham, Birmingham, United Kingdom.President of Vascular Access Society of Britain and Ireland.ReDVA programme lead.

Tamara JemcovConsultant Nephrologist and Head of Department, Department of Nephrology, Clinical Hospital Centre Zemun, Belgrade, SerbiaClinical Teaching Assistant in Internal Medicine, School of Medicine, University of Belgrade, Serbia.

Stephanie KershawDialysis Access Nurse Specialist at the Norfolk and Norwich University Hospital, Norfolk, United Kingdom.

Aurangzaib KhawajaFellow Vascular Surgeon, Specialist Doctor, RedVA fellow, Department of Renal Transplantation and Access and renal transplant Surgery, Queen Elisabeth Hospital, University Hospitals Birmingham, West Midlands Deanery, Birmingham, United Kingdom.

Mick KumwendaConsultant Nephrologist, Glan Clwyd Hospital, Rhyl, Denbighshire, United Kingdom.

Laura LabriolaConsultant Nephrologist, Cliniques universitaires Saint-Luc, Brussels, Belgium.

Carlo LomonteChief of the Division of Nephrology, Miulli General Hospital, Acquaviva delle Fonti, Italy.

Marko MalovrhProfessor of Internal Medicine, Specialist of Internal Medicine and Nephrology, Medical Centre Ljubljana, Ljubljana, Slovenia.

Anna Marti i MonrosNephrology Nurse, Hospital General Universitario, Valencia, Spain.

Shona Matthew NIHR Unit Manager and Researcher, School of Medicine, University of Dundee, Scotland.

Damian McGroganResearch Fellow Vascular Access Surgery, West Midlands Deanery, Birmingham, United Kingdom.

Torsten MeyerConsultant Nephrologist, Department of Medicine V (Nephrology and Hypertension), City Hospital Braunschweig, Germany.

of 226

Sotirios MikrosConsultant Nephrologist, Associate Hospital Manager, Thriassion General Hospital, Athens, Greece

Nils PlankenCardiovascular Radiologist, Department of Radiology and Nuclear Medicine, Amsterdam University Medical Center, Amsterdam, The Netherlands.

Steve PowellChief Diagnostic Officer, Rutherford Diagnostics, Newport, United Kingdom.

Ramon Roca-TeySenior Consultant in Nephrology, Hospital de Mollet, Fundació Sanitària Mollet, Barcelona, Spain.

Rose RossHead Of Service, Clinical Specialist Vascular and Respiratory Services, Ninewells Hospital Scotland, United Kingdom.

Jan TordoirAssociate Professor of Surgery at the department of Vascular Surgery of the Maastricht University Medical Centre, Maastricht, The Netherlands.

Max TroxlerMax Troxler, Consultant Vascular Surgeon, Leeds Vascular Institute, Leeds Teaching Hospitals Trust, United Kingdom.

Raymond VanholderProfessor Emeritus Nephrology, Ghent University Hospital, Ghent, Belgium.

Frank VermassenHead of Department of Thoracic and Vascular Surgery, Ghent University Hospital, Ghent, Belgium.

Gunilla WelanderConsultant Nephrologist Centralsjukhuset Karlstad, Sweden. Member of board Swedish Renal Registry.

Teun WilminkConsultant Vascular Surgeon at the Heart of England NHS foundation Trust in Birmingham, United Kingdom.

ERBP Methods support team

Davide BolignanoNephrologist and Clinical Researcher, Institute of Clinical Physiology of the Italian National Council of Research, Reggio Calabria, Italy.

Christiane DrechslerNephrologist, University of Würzburg, Würzburg, Germany.

Jonathan FoxNephrologist, Chair of ERBP, University of Glasgow, Glasgow, United Kingdom.

Maria HallerResident Nephrology, Ordensklinikum Linz Elisabethinen, Linz, Austria.

of 226

Muguet KoobasiInformation Specialist, ERBP, guideline development body of ERA-EDTA, London, United Kingdom.

Evi Nagler Consultant Nephrologist, Vice-Chair of ERBP, Ghent University Hospital, Ghent, Belgium.

Ionut NistorConsultant Nephrologist, C. I. Parhon Hospital, Grigore T. Popa, University of Medicine and Pharmacy, Iasi, Romania.

Wim Van Biesen Consultant Nephrologist, Ghent University Hospital, Belgium.

Sabine van der VeerResearch Fellow, Centre for Health Informatics, University of Manchester, Manchester, United Kingdom.

6. Conflict of interest policy and declaration of interest formsWe required all participants in the Guideline Development Group to fill out a detailed ‘Declaration of Interest Statement’ including all current and future conflicts of interest as well as past interest restricted to the two years before joining the guideline development process. Declaration of interest statements of individual Guideline Development Group members are enclosed in Supplement 2 (Supplement 2 - Declaration of interest statements).

Because it was felt that excluding every individual with some degree of possible conflict of interest would make assembling a guideline development group impossible, we allowed members of the guideline development group to have past financial and/or intellectual conflicts of interest. We did not attach any consequences to the stated interests, but rather insisted on transparency. All members of the guideline development group could participate in all the discussions and have equal weight in formulating the statements. All were allowed equal involvement in data extraction and writing the rationales.

of 226

7. Purpose and Scope of this guideline

7.1. Topic selectionThe process of identifying and prioritising all relevant treatment decisions along the vascular access care pathway (vascular access ‘topics’) comprised three phases [5]. In phase 0, we created a preliminary list of topics based on a literature review, and input from a multidisciplinary expert group. Its members did not necessarily participate in later stages of the guideline development process. The group consisted of two kidney patients, two nephrologists, a renal nurse, two surgeons, and a radiologist. In phase 1, an international panel of 85 kidney patients, 687 nephrologists, 194 nurses, and 140 surgeons/radiologists rated the priority of these topics on a 5-point Likert scale through an online survey, and suggested additional topics to complement the preliminary list. The additional topics were prioritised in phase 2, by rating 42 vascular access related topics on a 5-point Likert scale in an electronic questionnaire.Details of the scoping procedures and its results have been published separately [5]. The group estimated it would be feasible to select 20 topics for further elaboration, and selected these from the list of 42 topics, guided by a preference for prioritizing topics that had been covered by the EBPG guideline and by the priority ratings they had received from patients and clinicians in the scoping procedure [2]. Although the group initially developed all 20 topics simultaneously, the guideline was later split into two parts for reasons of feasibility.

7.2. Why was this guideline produced?The purpose of this CPG was to provide guidance on the management and preservation of AV fistulas and grafts for haemodialysis. It was designed to provide information and assist decision-making related to this topic. It was not intended to prescribe a standard of care and should not be construed as such. It should also not be interpreted as prescribing an exclusive course of management.This CPG was developed by ERBP, the guidance body of the European Renal Association – European Dialysis and Transplant Association (ERA-EDTA) and is a collaborative effort of various stakeholders within the field, including representatives of the Vascular Access Society (VAS), nephrologists, vascular access surgeons, radiologists, dialysis nurses, researchers, patients and their carers.

All these stakeholders agreed there was a need for up to date guidance on vascular access management. The current document is an update of EBPG guideline for haemodialysis published in 2007 [2]. An attempt to adhere to increasingly stringent guideline development methodology has required certain sacrifices in terms of scope. As a result, the current document does not necessarily cover the same topics as the previous version. Some are shared, but some were archived in favour of new questions prioritised by both healthcare providers and the people they care for [5].

7.3. Who is this guideline for?This guideline aims to support clinical decision-making for any health care professional treating or caring for haemodialysis vascular access, including: nephrologists, vascular access surgeons, radiologists, dialysis nurses, and pharmacists dealing with

vascular access in both outpatient and in-hospital settings general practitioners, internists and surgeons not directly practicing in the nephrology or dialysis

access field but who may be confronted indirectly with haemodialysis access issues and systems.

The guideline is also developed for: policymakers for informing standards of care at a national and international level haemodialysis patients to improve their views on what dialysis access is about and how they could

participate in its maintenance and preservation.

of 226

7.4. What is this guideline about? This CPG covers aspects related to the vascular access that are necessary for successful long-term haemodialysis. A rigorous multi-layered phased selection procedure drove identification of the specific clinical questions this guideline aims to answer [5].

7.4.1. Population

The guideline covers aspects related to the vascular access that is necessary for successful chronic haemodialysis in adults of all ages with end-stage kidney disease. It does not cover vascular access in children, because the guideline development group felt essential differences exist between the two patient groups, requiring a targeted guideline development process. Not only would priorities for guideline development likely differ, interventions would have an appreciably different risk-benefit balance, and require exploration of lower level evidence generated specifically in children, which would be beyond the limits of our resources available at present.

7.4.2. Conditions

This guideline covers aspects related to maturation and maintenance of AV fistulas and grafts used in long-term haemodialysis. It specifically deals with interventions for promoting maturation of the AV access; perioperative antibiotic therapy for preventing AV access infection; timing of first cannulation; cannulation techniques and needle types; medical treatments for long-term AV access patency; AV access surveillance and pre-emptive intervention; and surgical and endovascular interventions for AV access thrombosis.

7.4.3. Healthcare setting

This guideline targets outpatient, in hospital and out of hospital haemodialysis unit settings dealing with adults who need to have an AV access for long-term haemodialysis.

7.4.4. Clinical management

This guideline deals with educational, pharmaceutical and interventional tools for promoting successful use of an AV access and interventions aimed at preventing failure of the vascular access, by the use of specific treatment strategies tailored to the underlying problem.This guideline covers the treatment for adults with acute or chronic vascular access problems, and strategies to prevent or treat these, regardless of the underlying cause of kidney disease, any pre-existing systemic vascular condition or any specific haemodialysis strategy.In line with the mission statement of ERBP, this guidance document intends to inform all involved stakeholders, and to stimulate shared decision-making. It also highlights topics for which additional research data are needed and offers suggestions for how research might be pursued [3].

of 226

8. Methods for guideline development

8.1. Establishment of the guideline development groupA steering committee consisting of the chair of ERBP at that time (Wim Van Biesen), selected members of ERBP’s methods support team (Christiane Drechsler, Maria Haller, Muguet Koobasi, Evi Nagler, and Sabine van der Veer), the co-chairs of the guideline development group appointed by ERBP’s Advisory Board (Maurizio Gallieni and Anna Marti I Monros), and selected content experts (Markus Hollenbeck, Nicholas Inston, Mick Kumwenda, Steve Powell, Jan Tordoir, Matthias Widmer) convened in May 2013 and February 2014, and decided on the composition of the guideline development group, taking into account the clinical and research expertise of the proposed candidates. The guideline development group consisted of content experts, which included individuals with expertise in vascular access. In its composition, the guideline development group aimed to be multidisciplinary, including nephrologists, surgeons, radiologists, researchers, a nurse and a patient. The ERBP methods support team provided methodological input and practical assistance throughout the guideline development process.

8.2. Developing clinical questions With the scope of the guideline as point of departure, the guidelines development group identified specific clinical research questions, for which a systematic review was conducted.

8.3. Development of review questionsThe methods support team assisted in developing review questions, i.e. framing the clinical questions into a searchable format, by applying the PICO procedure. This required careful specification of the patient group (P), the intervention (I), the comparator (C) and the outcomes (O) for intervention questions and the patient group, index tests, reference standards and target conditions for questions of diagnostic test accuracy [6]. For each question, the guideline development group agreed upon explicit review question criteria including study design features. (Supplement 3 for Detailed Review Questions and PICO tables).

8.4. Assessment of the relative importance of the outcomesFor each intervention question, the guideline development group compiled a list of outcomes, reflecting both benefits and harms of alternative management strategies. The guideline development group ranked the outcomes as critical, highly or moderately important according to their relative importance in the decision-making process. As such, patient-important health outcomes such as patient survival and quality of life, as well as permanent loss of the vascular access, were considered critical. Outcomes such as AV access thrombosis, infections, hospitalisations, and temporary central venous haemodialysis catheter use, were considered highly important, but less important than the critically important clinical outcomes (Table 1).

8.5. Target population perspectivesEfforts were made to capture the target population’s perspectives by adopting three strategies. To identify and prioritise all relevant treatment decisions along the vascular access care pathway (vascular access ‘topics’), we conducted an extensive survey in three phases, including participants with kidney disease in every phase of the process [5]. We ultimately elicited responses from 85 patients, residing in Austria (15%), Belgium (24%), Spain (14%), the Netherlands (29%) and the United Kingdom (18%). Second, one of the guideline development group members, a researcher in the field of vascular access, actively involved with the evidence review and guideline development process, had previously been treated with chronic haemodialysis and, as such, was very familiar with many of the challenges faced by people with end-stage kidney disease (ESKD).

of 226

Third, ERBP has a permanent patient representative on its board. Although he was not included in the guideline development group or in the evidence review process, drafts of the guideline document were sent out for his review and his comments were considered in revising and drafting the final document.

8.6. Searching for evidence

8.6.1. Sources and search strategy

We used a hierarchical strategy, whereby the ERBP methods support team first searched the Cochrane Database of Systematic Reviews (up to April 2018) for reviews that were either up-to-date or could be updated by our review team. If such a review did not exist, the methods support team subsequently searched DARE (up to April 2018), CENTRAL (up to April 2018) for other systematic reviews and randomised trials respectively. If no randomised trials were available, an additional search was conducted in MEDLINE (up to April 2018) for identifying non-randomised studies that fit the inclusion criteria set for each research question. Search strategies combined subject headings and text words for the patient population and intervention. The detailed search strategies and dates are available from Supplement 4 Search strategies.

8.6.2. Study selection

We used the Early Reference Organisation Software - EROS (http://www.eros-systematic-review.org) to organize the initial step of screening and selection of papers. The title and abstract of all papers retrieved by the search were made available to those responsible for screening through this system. For each research question, two guideline development group members independently screened all titles and abstracts. Abstracts that did not meet the inclusion criteria were discarded. Any discrepancies at this stage were resolved by group consensus. In a second step, the methods support team retrieved full texts of potentially relevant studies and two mutually independent reviewers examined them for eligibility, according to pre-set eligibility criteria and independently of each other. Any discrepancies were resolved by consensus. If no consensus could be reached, the disagreement was settled by group arbitrage.

The flow diagram depicting the paper selection process for each research question is presented in Supplement 5 Study selection and flow diagrams.

8.6.3. Data extraction and critical appraisal of individual studies

For each included study, we collected relevant information on design, conduct, and relevant results through standardised data extraction forms. These forms were developed in Salesforce®, a customer relationship platform, customized to fit our needs. We introduced closed questions with drop-down answer lists whenever possible to improve data quality. The tool allowed automatic collection of the entered data into a database and semi-automatic cross-checking of the data independently entered by two reviewers. It also facilitated the generation of the summary evidence tables directly from the collated dataset. As no data required manual copying into a different format after it had been entered by the reviewer, we believed this would reduce error in handling of the data. For each question, two reviewers extracted all data independently of each other. The Methods Support Team produced tables displaying the extracted data for both reviewers by question. A member of the Methods support team checked all the data and any discrepancies were discussed and resolved by consensus, and if no consensus could be reached, disagreements were resolved by an independent referee. From these tables, we produced merged consensus evidence tables for informing the recommendations. The summary evidence tables are available from Appendix 6 Summary evidence tables.Risk of bias of the included studies was evaluated using various validated checklists, as recommended by the Cochrane Collaboration. These were AMSTAR for Systematic Reviews [7], the Cochrane Risk of Bias tool 2.0 for randomised controlled trials [8], the Newcastle Ottawa scale for cohort and case-control studies [9] and QUADAS for diagnostic test accuracy studies [10]. Data were compiled

of 226

centrally by the ERBP methods support team.

8.7. Rating the quality of the evidence for each outcome across Studies The evidence for outcomes on therapeutic interventions from the included systematic reviews of randomised trials was assessed using the ‘Grading of Recommendations Assessment, Development and Evaluation (GRADE) toolbox’ developed by the international GRADE working group [11]. In accordance with GRADE, the guideline development group initially categorized the quality of the evidence for each outcome as high if it originated predominantly from randomised controlled trials and low if it originated from observational data. The quality of the evidence was subsequently downgraded one or two levels if results from individual studies were at serious or very serious risk of bias, there were serious inconsistencies in the results across studies, the evidence was indirect, the data were sparse or imprecise or publication bias was thought to be likely. If evidence arose from observational data, but effect sizes were large, there was evidence of a dose-response gradient, or all plausible confounding would either reduce a demonstrated effect or suggest a spurious effect when results showed no effect, the quality of the evidence would be upgraded (Table 2). Uncontrolled case-series and case-reports automatically received downgrading from ‘low’ to ‘very low’ level of evidence for risk of bias, so that no further reasons for downgrading were checked. By repeating this procedure, an overall quality of evidence for each outcome and each intervention was obtained. See Table 3 for the list of definitions.

8.8. Formulating statements and grading recommendations

8.8.1. Recommendations

After the summary tables had been prepared and the evidence assessed, recommendations were formulated and graded. Recommendations can be in favour or against of a certain strategy. The guideline development group drafted the recommendations based on their interpretation of the available evidence. Judgements around four key factors determined the strength of a recommendation: the balance between desirable and undesirable consequences of alternative therapeutic or diagnostic strategies, the quality of the evidence, the variability in values and preferences. Formal decision or cost analysis was not conducted. In accordance to GRADE, the strength of the recommendations was classified as strong, coded ‘1’ or weak, coded ‘2’ (Table 4) [11]. The strength of a recommendation is determined not only by the certainty of evidence, but also by other, often complex judgments regarding the size of the net medical benefit, values and preferences, and costs. Individual statements were made and discussed to reach group consensus.

8.8.2. Advice for clinical practice

An additional category of ungraded statements was used for areas where formal evidence was not sought, and statements were based on common sense, or expert experience alone. They were termed ‘advice for clinical practice’ to differentiate them from graded recommendations and do not hold an indicator for the quality of the evidence. Advice for clinical practice is only for improving practical implementation. It contains some elaboration on one of the statements, clarifying how the statement can be implemented in clinical practice. The ungraded statements were generally written as simple declarative statements but were not meant to be stronger than level 1 or 2 recommendations.

8.9. Writing rationaleRecommendations and advice for clinical practice for each of the clinical questions were collated in separate chapters structured according to a specific format. Each question resulted in one or more specific boxed statements. Within each recommendation the strength was indicated as level 1 or level

of 226

2, and the quality of the supporting evidence as A, B, C or D as prescribed by the GRADE methodology (Table 3 and 4) [11]. These statements are followed by advice for clinical practice and the rationale. The rationale contains a brief section with relevant background and justification of the topic, followed by a short narrative review of the evidence, and finally a justification of how the evidence was translated in the recommendations made. When areas of uncertainty were identified, the guideline development group considered making suggestions for future research based on the importance to patients or the public, and on ethical and technical feasibility. Finally, each chapter provides an overview of recommendations made by other guideline bodies. The list is not meant to be exhaustive, but rather intended to provide a concise overview of other recommendations made. Any grading was reprinted as reported in the respective clinical practice guideline. These may and do for some organisations differ from the GRADE system used for the ERBP guideline development process. The reader is referred to the original publication for further details regarding the grading system for coding individual recommendations.

8.10. Internal and external review

8.10.1. Internal review

Both the Vascular Access Society (VAS) and ERBP nominated experts in vascular access and members of their governance bodies. Internal referees were asked to complete a standardised internal review survey online (Supplement 7 Internal Review). Referees were asked whether statements were clear, implementable and to what extent they agreed with the content on a scale from one to five. These scores were averaged and color-coded from red (1) to green (5) to help visualise problematic areas. In addition, internal reviewers were asked to comment on the statements and the rationale within free text-fields. All these comments and suggestions were discussed with the guideline development. For each comment or suggestion, the guideline development group evaluated if the statement needed to be adapted, again considering the balance between desirable and undesirable consequences of the alternative management strategies, the quality of the evidence, and the variability in values and preferences.

8.10.2. External review

The draft guideline was posted on the ERBP website and the public invited to comment using the same standardised review form as was used for the internal review process. Anyone was invited to comment, but reviewers had to provide basic information to identify their background or stakeholder position. The public consultation period lasted for four weeks to ensure adequate time for responses. In addition to public consultation, the guideline was sent to the Council of ERA-EDTA, the Vascular Access Society (VAS), Kidney Diseases Initiative Global Outcomes (KDIGO), Kidney Health Australia – Caring for Australasians with Renal Impairment (KHA-CARI), and the National Institute of Clinical Excellence (NICE), the European Dialysis and Transplant Nurses Association/European Renal Care Association (EDTNA/ERCA), and the European Society of Vascular Surgery (ESVS) for review. Referees used the same standardised review form as for the internal review process. All comments were summarised by the ERBP Methods Support Team and provided to the guideline group chair to determine the appropriate course of action as part of finalising the guidelines. As part of the final approval process, the ERBP Chair and Co-chair ensured that all comments had been appropriately addressed.

8.11. Timeline and procedure for updating the guidelineERBP aims to update its clinical practice guidelines at least every five years. New evidence requiring additional recommendations or changes to existing statements could instigate an earlier update.

of 226

At least every five years, the ERBP methods support team will aim to update its literature searches. Relevant studies will be identified, and their data extracted using the same procedure as for the initial guideline. During a one-day meeting, the guideline development group will decide whether the original statements require updating. ERBP will aim to publish an updated version of the guideline online.

8.12. FundingActivities of ERBP and its methods support team are supervised by an advisory board (see www.european-renal-best-practice.org for details and declaration of interests). ERBP is a working group of ERA-EDTA. The Council of ERA-EDTA approves and provides the annual budget based on a proposition made by the chair of ERBP. ERA-EDTA is partly funded by industry, but its council is not involved with and does not interfere with topic choice, question development or any other part of the guideline development process. Neither the societies nor the guideline development group received any funds directly from industry to produce this guideline.

of 226

9. Chapter 1. Medical treatments for promoting AV fistula maturation

9.1. Recommendations

We suggest any decision to give aspirin, ticlopidine, or clopidogrel in adults with end-stage kidney disease during the first two months after AV fistula creation for the sole purpose of improving maturation, must balance a reduction in thrombosis against uncertain effects on maturation and bleeding. (2C)

We suggest any decision to give perioperative heparin in adults with end-stage kidney disease during AV fistula creation, must balance an increase AV fistula patency at one month against an important increase in bleeding complications. (2C)

We suggest any decision to apply far infrared therapy in adults with end-stage kidney disease during the first three months after AV fistula creation, must balance a possible reduction in thrombosis against uncertain effects on maturation and bleeding. (2C)

There are insufficient RCT data to make a recommendation for ticagrelor, prasugrel, dipyridamole, sulphinpyrazone, warfarin or other oral anticoagulants, fish oil, statins, vonapanitase, glycerine trinitrate, iontophorectic injection of Salvia miltiorrhiza, or prednisolone for improving AV fistula maturation in adults with end-stage kidney disease. (-D)

Advice for clinical practice Do not stop mono-antiplatelet treatment in adults undergoing AV access creation.

Rationale

Background Non-maturation is defined as a process leading to a newly created AV access which cannot be used for haemodialysis; it does not apply to AV grafts. Non-maturation may cause various problems, such as a need for re-intervention or for a temporary central venous haemodialysis catheter to be inserted. An AV fistula may fail because of thrombosis, or because of the feeding artery or the draining vein failing to enlarge. Medications that influence these processes could result in improvement of maturation, provided their adverse effects do not counterbalance their benefits either locally or systemically. For instance, anticoagulant and antithrombotic agents may prevent clotting, but they may also cause bleeding. Vasoactive agents may prevent vasospasm, stimulate vasodilation, and increase blood flow in the newly created AV fistula, but they may also lower systemic blood pressure. As maturation problems in the first months may result from different pathophysiological mechanisms than patency problems in the longer term, the two issues are discussed in two different chapters of this guideline. The current chapter specifically covers AV fistula maturation. Chapter 8 covers long-term patency of AV fistulas and AV grafts.

Summary of the evidence (Supplement 3 | Review Questions – PICOM Format - Chapter 1)(Supplement 4| Search Strategies - Chapter 1)(Supplement 5| Study selection flow diagrams - Chapter 1)(Supplement 6| Summary evidence tables - Chapter 1)

We identified seven systematic reviews of RCTs assessing benefits and harms of various medical adjuvant treatments to increase overall patency of AV fistulas and AV grafts [12-18]. All these reviews were judged to be of moderate to high quality with AMSTAR scores of 8 to 10/11. The reviews included studies measuring maturation outcomes after six to 12 weeks, and patency outcomes

of 226

measured several months later. Unfortunately, the meta-analyses did not separate studies reporting maturation outcomes from studies reporting longer term patency outcomes. The next paragraph describes the nature and content of the included systematic reviews that were used to identify relevant randomised trials. Based on group consensus, for this chapter, we chose to only consider RCTs and meta-analyses measuring patency outcomes before or at 12 weeks, as an arbitrary cut-off to distinguish maturation from long-term patency; and only in those studies assessing AV fistulas.

The first was a Cochrane systematic review with content assessed as being up-to-date to 23 March 2015 [13]. It included 15 RCTs comprising 2230 participants at the time of access creation. Seven trials included patients with an AV fistula, six with an AV graft, and two with either an AV fistula or an AV graft. All but one enrolled participants at the time of AV access creation [19]. Tested medical interventions included aspirin, ticlopidine, dipyridamole, dipyridamole plus aspirin, warfarin, fish oil, clopidogrel, sulphinpyrazone, and human type I pancreatic elastase (vonapanitase). Studies mostly included participants of all ages and follow-up ranged from 1 to 18 months after AV access insertion. Most studies only assessed AV access thrombosis as primary outcome of interest. The second was also a Cochrane systematic review, which specifically covered anti-platelet agents to prevent vascular access failure and other outcomes in people with chronic kidney disease. Its content was assessed as being up-to-date to 24 January 2011 [14]. The review included 12 RCTs assessing the effect of antiplatelet agents in a newly created AV access, and another nine in an already functioning AV access. Nine of these were also included in the first review [13]. Of the three remaining studies, the largest had been excluded [20]. The RCT tested dipyridamole plus aspirin versus placebo but included participants in whom aspirin was started prior to study inclusion and did not require discontinuation in the placebo group. The other two did not contribute to any of the meta-analyses of interest [21, 22]. Medical interventions included aspirin, ticlopidine, dipyridamole, dipyridamole plus aspirin, clopidogrel, and sulphinpyrazone [14]. The Cochrane review resulted in a derivative review only assessing vascular access outcomes [12].Four other systematic reviews each assessed a specific treatment such as fish oil [15], far infrared therapy [16, 17], or intraoperative anticoagulation during AV access formation [18]. In addition to these reviews, we identified six RCTs, published after 2013, and not included in any of the considered systematic reviews, assessing various adjuvant medical treatments for improving AV access patency [23-28].

Antiplatelet agentsPalmer and co-workers found that, overall, antiplatelet agents seemed to reduce AV fistula thrombosis at eight weeks, although the certainty of the evidence was compromised by risk of bias in the underlying studies and serious imprecision, with a total sample size below the optimal information size (5 RCTs; n=1005; RR 0.43 – 95%CI 0.26 to 0.73; I²=25%) [12]. Effects on AV fistula maturation failure were uncertain (2 RCTs; n=794; RR 0.57 – 95%CI 0.13 to 2.51). Definitions for AV fistula maturation varied widely.

AspirinOne RCT compared 500 mg of aspirin versus placebo given the day before, until 28 days after, construction of a Brescia-Cimino fistula in 92 patients [29]. Aspirin reduced AV fistula thrombosis at one month by 81% (n=92; RR 0.19 – 95%CI 0.04 to 0.81). The sample size was low and risk of bias unclear. Numerically more people complained of gastric pain and epistaxis in those taking aspirin, but the confidence interval was wide (11% versus 4%, 95%CI 4% fewer to 18% more). The risk of gastro-intestinal bleeding or wound haematoma was the same in both groups (4%).

TiclopidineThree RCTs compared ticlopidine 250 mg twice daily versus placebo for one month in 339 patients undergoing AV fistula creation [13]. Combining results, it appeared ticlopidine may reduce AV access thrombosis, but the certainty of the evidence was low due to risk of bias in the primary studies and serious imprecision (3 RCTs; n=339; OR 0.45 – 95%CI 0.24 to 0.85; I²=20%). There was no clear information about adverse events.

of 226

Clopidogrel75 mg of clopidogrel was compared versus placebo in two RCTs, indicating it may reduce AV fistula thrombosis up to six weeks after AV fistula creation (2 RCTs, n=959; OR 0.40 – 95%CI 0.13 to 1.19; I²=54%) [13]. The certainty of the evidence was low, however, due to risk of bias in both studies, and serious imprecision, with the confidence interval spanning the line of no effect. Both studies indicated a similar proportion of bleeding events and mortality.

We identified an additional RCT including 96 participants randomised to either clopidogrel plus a prostacyclin analogue (epoprostenol) or placebo for 7 days before and up to one year after surgery [23]. The investigators described an important improvement in fistula maturation - in this study defined as a blood flow >300 mL/min or velocity > 70 cm/sec - for those taking dual antiplatelet treatment (87% versus 67%), but certainty of the evidence was low due to unclear denominators. The investigators reported no bleeds, but external validity of the results was considered low because of stringent exclusion criteria. Only 25% of all people set to undergo AV fistula creation were finally enrolled.

Dipyridamole and SulphinpyrazoneThere were no studies assessing outcomes at 12 weeks for dipyridamole or sulphinpyrazone.

AnticoagulantsWarfarin or other oral anticoagulantsThere were no studies assessing outcomes at 12 weeks for warfarin or other oral anticoagulants.

Perioperative anticoagulantsWe identified one systematic review on systemic intra-operative anticoagulation during AV access formation [18]. The review included three randomised trials which used systemic heparin during AV fistula creation and found the intervention increased AV fistula patency at 6 weeks, but quality of the evidence was low due to high risk of bias in the underlying studies and large imprecision of the summary effect estimate (3 RCTs; RR=0.57 – 95%CI 0.33 to 0.97). Importantly, systemic heparin increased bleeding events in all access trails including an additional RCT including both AV fistulas and grafts (4 RCTs; n=411; RR 7.18 – 95%CI 2.40 to 21.40). Another RCT assessed the combined use of heparin and anisodamine (an anticholinergic and α1 adrenergic receptor antagonist used in the treatment of acute circulatory shock in China) given immediately after AV fistula creation and found it reduced the risk of early thrombosis in AV fistulas compared with placebo or heparin alone, but the study was at high risk of bias [30].

OtherFish oil – omega 3 polyunsaturated fatty acidsWe identified a systematic review comparing fish oil versus placebo or no treatment, with content assessed as being up-to-date to January 2017 [15]. In addition to two RCTs identified by Tanner and colleagues, this review included four additional RCTs assessing dosages ranging from 3 g three times weekly to 6 g daily. Only one study assessed the effect in AV fistulas, the others in AV grafts [31]. Outcomes were measured at 12 months, there were no data on outcomes measured within the prespecified time period of 12 weeks.

StatinsThere were no studies assessing outcomes at 12 weeks for statins.

Vonapanitase – recombinant type I pancreatic elastase Two RCTs assessed the effect of recombinant type I pancreatic elastase applied directly to the adventitia of the AV vessels at the time of AV fistula creation [13]. For AV fistulas, there was no difference in unassisted maturation at six weeks regardless of the definition of maturation. At three months, more AV fistulas had matured in the group that had received vonapanitase, but a large loss to follow up and multiple testing issues reduced the certainty of the evidence.Both studies listed several adverse events including local symptoms secondary to the creation of a new

of 226

AV fistula, but according to the study authors there were no significant differences between placebo and recombinant type I pancreatic elastase-treated participants. We found an additional trial that randomised 313 people to locally dripped vonapanitase or placebo at the time of AV fistula creation. The report provided no outcome data for maturation [24].

Far infrared therapyWe found one systematic review on far infrared treatment for increasing the patency of AV fistulas, with content assessed as being up-to-date to January 2017 [17]. The review included 21 RCTs comprising 1899 participants at the time of AV access creation. Although not clearly reported, it appears all studies included participants with an AV fistula. Investigators used different techniques for delivering of far infrared rays; most commonly a device generating wavelengths between 5 and 25 mm, set at a height of 20 cm above the AV fistula with a treatment time of 40 min during each haemodialysis session. Most studies assessed blood volume, vessel diameter, or primary patency at various time points, up to one year after fistula creation. We awarded the review an 8/11 score on the AMSTAR checklist, limited by the absence of a registered protocol, unclear search methods, and incomplete reporting of excluded studies. Eight studies assessed outcomes earlier than or at three months after AV fistula creation, but only four could be included in meta-analysis. Overall, far infrared therapy modestly increased blood flow, although the certainty of the evidence was compromised by the small number of studies and research groups these data were derived from, the risk of bias in the underlying studies, and the heterogeneity in the results (3 RCTs; n=328; mean difference (MD) 57.2– 95%CI 9.1 to 105.2; I²=93%). In addition, two trials comprising 180 participants showed results for AV fistula occlusion rates. Overall, therapy with far infrared radiation decreased AV fistula occlusion rates (2 RCTs; n=180; RR 0.29 – 95%CI 0.06 to 1.35, I²=0%). There was no evidence of heterogeneity in these trials.

CholecalciferolOne RCT randomised 52 participants to receive either 200,000 IU oral cholecalciferol or matching placebo [25]. High dose cholecalciferol had uncertain effects on AV fistula maturation at six months (1 RCT; n=52; RR 0.79 – 95%CI 0.45 to 1.38).

Glyceryl trinitrateGlyceryl trinitrate was tested in a placebo-controlled clinical trial including 200 participants [26]. An active or a placebo patch was applied 5 cm proximal to the AV anastomosis immediately after completing the surgical procedure and left in place for 24 hours. The trial provided moderate certainty evidence for no meaningful difference in patency at six weeks or change in venous diameter as surrogate for maturation. There were no major differences in side effects resulting in similar numbers dropping out in both groups, but no formal analysis was presented.

Iontophoretic intradermal injection of Salvia miltiorrhizaWe found one Chinese trial randomising 20 participants who had undergone radio-cephalic AV fistula creation to iontophoretic intradermal injection (a technique that uses low-level current to drive charged compounds across the skin) of Salvia miltiorrhiza, a Chinese root that contains salvianolic acid B as a potentially vasculo-protective and anti-inflammatory agent [27]. It found the experimental treatment increased the number of successfully matured AV fistulas by 35% at one month (1 RCT; n=20; RR 1.07 – 95%CI 1.06 to 2.73). We considered the evidence of low certainty because of the sample size on one hand, and concerns around changes in the primary outcome on the other hand – originally to be measured at six months rather than one month.

PrednisoloneFinally, we identified a protocol for an RCT set to assess the effect on radio-cephalic AV fistula maturation of intravenous liposomal prednisolone 150 mg given twice after surgery (day 1 and 14) [28]. To the best of our knowledge, results of this study were not yet available upon publication of the guideline.

of 226

Translation of the evidence into recommendationsInterpreting the available data in the context of maturation is challenging for various reasons. Most studies assessing antiplatelet agents report on short term vascular access thrombosis rather than successful dialysis. This is problematic as a reduction in AV fistula thrombosis does not necessarily translate into improved maturation. It is true that fistula thrombosis precludes successful use of the AV access for dialysis, but if the current treatments, predominantly, aimed at reducing platelet aggregation and coagulation, increase the risk of bleeding, a local hematoma may cause irremediable loss of the access even before it has ever been used. In addition, access thrombosis may be treated using endovascular or surgical techniques, and antiplatelet agents have uncertain effects on reducing interventions for assisting maturation. Authors use different definitions for the concept of AV fistula maturation and that too complicates interpretation of the data. Some investigators treat maturation as a pre-cannulation outcome based on surrogate measures of vessel diameter and blood flow. Whether or not the AV fistula is successfully used for dialysis later on is often ignored. The guideline development group judged that an improvement in maturation using pre-cannulation definitions would not be enough to issue a supporting recommendation. Last, many studies report primary unassisted patency after one year and do not distinguish between the maturation phase and long-term patency of a matured AV fistula. As harmful effects of treatments may change over time, differences in primary unassisted patency may well be non-proportional too. Put otherwise, what benefits the maturation process may be different from what benefits the matured AV fistula.The guideline development group felt that for a positive recommendation, interventions had to improve successful use of the AV access. We judged that in the absence of evidence for a positive effect of successful cannulation, evidence for an effect on intermediate outcomes such as AV access thrombosis would not be enough to advocate treatment. But rather than formulating a neutral statement the group also wanted to highlight existing ambiguity by communicating the items to be weighed in decision-making. After initial recommendations were drafted, the group decided to add a statement advising not to stop antiplatelet treatment in adults already treated with antiplatelet agents for other reasons. Although this chapter did not directly aim to answer that question, it was felt current evidence supporting continuation of antiplatelet treatment in adults undergoing non-cardiac surgery would tip the uncertain benefit for maturation in favour of continuing treatment [32].

Other guidelines on this topicESVS [33] We suggest individualizing the indication of anti-platelet agents to prevent thrombosis of native AV fistulas, given that, although a reduction in the risk of thrombosis was demonstrated, the adverse effects have not been well studied.

GEMAV [34]We suggest that antiplatelet therapy for thrombosis prophylaxis of native arteriovenous fistula be indicated on a case-by-case basis, because although it shows a decrease in the risk of thrombosis, we consider that adverse effects have not been studied with sufficient accuracy.

We suggest that antithrombotic prophylaxis not be used in patients with prosthetic arteriovenous fistula, because there is no benefit in preventing thrombosis and adverse effects have not been studied with sufficient accuracy

CSN, KDIGO, NKF-KDOQI, KHA-CARI, or NICE provide no current recommendation on this topic.

Suggestions for future researchLarge heterogeneity in reported trial outcomes limit our ability to interpret the evidence correctly. Therefore, we welcome initiatives like SONG-HD, which aim to develop standardised outcomes in vascular access [4]. Implementing consensus outcomes in future trials will be key for improving the

of 226

ability of systematic reviewers and guideline developers to better assess the benefits and harms of proposed treatments. Routinely collected data on vascular outcome data in registries might be of benefit to assess strategies and practices.

Most studies assess AV access thrombosis rather than AV access maturation or the ability to perform dialysis using a newly created AV access. In the presence of an adverse effect which could counter the intermediate outcome, a trial including maturation would be informative. Such a trial would likely capture other causes of maturation failure, which are currently not taken into consideration.

of 226

10. Chapter 2. Surgical and endovascular interventions for promoting AV fistula maturation


We suggest using regional block anaesthesia rather than local anaesthesia for AV fistula creation in adults with end-stage kidney disease. (2C)

We suggest there is insufficient evidence to support end-of-vein to side-of-artery over side-of-vein to side-of-artery anastomosis for AV fistula creation in adults with end-stage kidney disease. (2C)

Advice for clinical practice -

Rationale

BackgroundNon-maturation is defined as a process leading to a newly created AV access which cannot be used for haemodialysis; it does not apply to AV grafts. Non-maturation may cause various problems, such as a need for re-intervention or for a temporary central venous haemodialysis catheter to be inserted. An AV fistula may fail because of thrombosis, or because of the feeding artery or the draining vein failing to enlarge.Physical interventions that influence these processes may result in improvement of maturation, provided their adverse effects do not counterbalance their benefits. Such interventions include different types of anaesthetic procedures, balloon assisted maturation, use of devices to connect the artery and the vein, ligation of accessory draining veins, dilatation of the main draining vein, or specific surgical techniques to make the anastomosis.

Summary of the evidence(Supplement 3| Review Questions – PICOM Format - Chapter 2)(Supplement 4| Search Strategies - Chapter 2)(Supplement 5| Study selection flow diagrams - Chapter 2)(Supplement 6| Summary evidence tables - Chapter 2)

Two systematic reviews [35, 36] and 16 RCTs assessing eight different interventions were identified [37-52].

One systematic review and meta-analysis included seven studies including 870 participants comparing the effects of regional versus local anaesthesia on several AV fistula outcomes [35]. With respect to maturation, the authors found that regional block anaesthesia produced an average 25 mL/min increase in blood flow (2 RCTs; n=78; MD 25.08 mL/min – 95%CI 19.40 to 30.76; I²=53%) and a 22% increase in primary unassisted patency (3 RCTs; n=246; RR 1.22 – 95%CI 1.08 to 1.37; I²=45%). Conversely, there were no important differences for AV fistula thrombosis and primary fistula failure. Certainty of the evidence was low, and the number of studies included in the pooled analysis limited.

Five RCTs compared regional block anaesthesia versus local site anaesthesia [37-41]. A first RCT included 126 participants and reported better outcomes with block anaesthesia for primary patency at three months (n=126; OR 3.3 – 95%CI 1.4 to 7.6) and immediate patency at time of hospital discharge (n=126; OR 4.3, 95% CI 1.5 to 15.5) [37]. In a second RCT including 60 patients, the authors found no meaningful differences in primary patency after brachial plexus block or local infiltration anaesthesia [38]. They did report wider diameters of the AV fistula vein and improved hemodynamic parameters including blood flow at six

of 226

months for brachial plexus block. However, no actual numbers and no further details on outcome assessment were provided, thereby limiting our ability to assess the certainty of these findings. A third RCT found no differences in thrombosis, maturation, time to maturation, or time to first cannulation after vertical infraclavicular block or local anaesthesia in 123 individuals undergoing creation of a first AV fistula [39]. However, the study did not report numeric data, and specific details on study design and conduct were missing. The remaining two studies had a smaller sample size, including 60 and 34 participants each, and described improvements in surrogate outcomes in the group that received regional block anaesthesia [40, 41]. The first of the two reported improved blood flow for patients in the infra-clavicular block group, but recorded similar access failures in both groups [40].

One RCT assessed whether stellate ganglion blockade with ropivacaine, given for seven days after surgery, improved outcomes after radio-cephalic fistula creation [42]. In comparison with usual care, stellate ganglion blockade reduced the frequency of thrombosis up to 24 hours after surgery from 8 to 2 events (n=50; denominators not reported). Time until maturation was also shorter in the intervention group (41 ± 7 days versus 77 ± 11 days). However, the number of participants in whom the outcome was assessed was not reported and definitions of maturation were variable. A recent systematic review and meta-analysis collecting data from seven studies and 986 participants found no differences in AV fistula patency at 3, 6, 12, and 24 months after an end-of-vein to side-of-artery or side-of-vein to side-of-artery surgical approach [36]. The overall outcome certainty of this review was very low due to the scarce number of studies included, most of which were observational, and non-randomised trials. Three RCTs, two of which were included in the review, compared a side-of-vein to side-of-artery anastomosis versus an end-of-vein to side-of-artery anastomosis [43-45]. A first RCT, including 71 participants, reported no difference in access patency and fistula thrombosis at three and nine months [43]. A second RCT enrolled 60 participants and found no difference in patent fistulas at six months [44]. A third enrolled 336 adults and found no meaningful differences in patency at six months [45]. No study addressed end-of-vein to end-of-artery anastomosis or other, newer techniques which are more rarely performed.

There were three RCTs that compared clips versus sutures in performing an end-to-side anastomosis for AV fistula creation [46-48]. All three found uncertain effects on primary patency or AV fistula maturation. One RCT compared ligation versus no ligation of the distal vein after side-to-side anastomosis in 60 patients [49]. The investigators reported 90% access survival at 90 days in the intervention group versus 83% in the control group (p>0.05). However, important aspects of study design and conduct were not reported, making any inference problematic.One RCT assessing the type of suture was identified [50]. The investigators compared continuous versus interrupted suture of the AV fistula in 40 participants. Access survival at 2 years was found to be similar in both groups.A small RCT comparing one-stage versus two-stage procedures for creating a brachio-basilic vein AV fistula, found no differences in primary and secondary patency, or in non-thrombotic post-operative complications [51]. Conversely, transposed brachio-basilic vein fistula maturation rate after one-stage procedures was lower compared to maturation rate after two-stage procedures (33% versus 100%; p=0.01), which led to premature termination of the trial. Serious limitations in the trial design, however, hamper meaningful inference. Finally, a small trial including 40 participants, compared balloon angioplasty of cephalic veins with a diameter ≤ 2 mm with hydrostatic dilatation and ligation of collateral veins [52]. The trial was considered at high risk of bias due to its small sample size, incomplete analysis, lack of outcome definitions, omitting of indications for interventions to prevent loss of access, and unclear blinding procedures.

Translation of the evidence into recommendations RCTs provided low to medium certainty evidence overall. However, lack of standardisation in

of 226

outcome reporting made inference particularly difficult. Five RCTs were found to provide evidence for block anaesthesia in comparison with local anaesthesia. Only one RCT was of considered at low risk of bias while the other four were considered at high risk of bias. All studies suggested benefit of using regional block anaesthesia, but there were several considerations that limited the strength of the recommendation to a discretionary one. First, the risk of bias in these studies was generally high and outcome data were mostly limited to surrogate outcomes. Second, switching from local anaesthesia to regional block anaesthesia could unwantedly complicate the procedure, may increase costs, and possibly even delay the access procedure. Third, the main advantage of regional block anaesthesia was felt to be vein dilation, which could also be achieved by other means, such as creating warm conditions.For the comparison of end-of-artery to side-of vein versus side-to-side anastomosis, there were two reports, which were considered at medium risk of bias, with available results insufficient to recommend one type of anastomosis over another, but equally insufficient to indicate equipoise between the two. Three reports were available on the comparison of clips versus sutures for AV fistula creation. Sample sizes were small, and the studies had important shortcomings, leaving important uncertainty as to the benefit of one technique over the other. Considering this uncertainty, the guideline development group felt technique choice should be left to the surgical team, based on experience and personal preference. It was felt any recommendation would confuse the end-user rather than clarify any ambiguity, such that no recommendation was formulated.The guideline group considered the other trials to be preliminary at best, providing limited basis for formulating a recommendation in either direction. Hence, they decided to refrain from making statements related to vein ligation, suture technique, angioplasty, or techniques for creating a brachio-basilic AV fistula. Other guidelines on this topicCSN, ESVS, GEMAV, KDIGO, NKF-KDOQI, KHA-CARI, or NICE provide no current recommendation on this topic.

Suggestions for future researchThe available reports do not resolve the uncertainty about long term effects and do so only incompletely with regards to side effects. New multicentre trials comparing regional block anaesthesia with other anaesthetic techniques would help strengthen the evidence base. Larger trials comparing different anastomotic techniques are also advocated. Standardization of outcomes and estimation of adequate sample sizes are needed.

of 226

11. Chapter 3. Surgical and endovascular interventions for non-maturing

AV fistulas


We suggest there is insufficient evidence to support open surgical over endovascular interventions as the preferred treatment for non-maturing AV fistulas in adults with end-stage kidney disease. (2D)

Advice for clinical practice Decisions on how to treat non-maturing AV fistulas are likely best based on local resources,

experience, and success rates. Institutions likely benefit from building a dedicated multidisciplinary vascular access team, with

clinical experience in various techniques available for non-maturing AV fistulas.

Rationale

BackgroundTo allow successful use of a newly created AV fistula, the outflow vein needs to enlarge sufficiently to allow insertion of one or two needles and sustain the blood flow required for adequate dialysis. Unfortunately, up to a quarter of newly constructed AV fistulas fail to mature [53]. The outflow vein may not enlarge for many reasons, some of which can be remedied by surgical or radiological endovascular interventions. If unsuccessful, however, such treatments may reduce quality of life by increasing the number of interventions, may increase workload and may inflate costs. They may also delay creation of an alternative permanent AV access.


No RCTs comparing the benefits or harms of surgical or radiological endovascular interventions versus one another or versus no treatment were identified.

A recent narrative review that included an attempt at comprehensively searching multiple databases, found 28 non-randomised uncontrolled studies recording clinical success, one-year primary patency, or one-year secondary patency of various surgical and radiological endovascular techniques (Table 5 and 6) [53].

Thirteen studies used balloon angioplasty – defined as any technique whereby a catheter is inserted in the AV fistula to dilate a stenotic vessel, mostly at the juxta-anastomotic or the draining vein. It led to clinical success in 43% to 97% of procedures with one-year primary unassisted patency in 28% to 72%, and one-year secondary patency of 68% to 97%. Rupture of the venous wall occurred in about 15% of cases, the majority of which could be managed with prolonged balloon inflation. In two studies, clinicians dilated diseased radial arteries feeding a radio-cephalic AV fistula, which resulted in one-year primary unassisted patency in 65% and 83% and one-year secondary patency of 86 % and 96% of cases. Balloon-assisted maturation is a technique whereby the vein of an AV fistula is subjected to staged, serial long-segment balloon angioplasty dilations (e.g. every two weeks), until it reaches the desired diameter and flow rate. It was investigated in four studies and was considered clinically successful in 55% to 89% of interventions, but none of the studies provided data for one-year primary unassisted or secondary patency. Adverse effects, including local bleeding, rupture and

of 226

thrombosis occurred frequently (>40% of procedures) [54]. Endovascular accessory vein obliteration is a procedure whereby a metal coil is inserted in a collateral competing vessel to increase blood flow through the AV fistula. It can be performed as an isolated intervention or in addition to balloon angioplasty, with variable results. Finally, balloon angioplasty has been conducted in combination with stent deployment in cases of stenosis of long venous segments with moderate one-year primary and secondary patency (65% and 72% respectively).

Surgical intervention usually involves a) proximal AV neo-anastomosis – defined as any technique whereby a new connection is created between the inflow artery and outflow vein; b) accessory vein ligation – defined as any procedure whereby collateral competing vessels are tied to increase blood flow through the AV fistula; or c) a combination of both. One-year patency in uncontrolled studies varies from 68% to 78% for primary patency and 85% to 95% for secondary patency. Several other techniques for aiding AV fistulas that are failing to mature have been experimented with during the last five years, including radial artery deviation and re-implantation, or placement of an internal or external anastomotic device. Individual studies are small and obtained success rates of similar magnitude to other surgical interventions.

A small retrospective study including 46 participants compared surgical proximal neo-anastomosis with endovascular intervention [55]. The group which had undergone surgical intervention had a cumulative AV fistula survival of 83%, compared to 43% for those who had undergone an endovascular treatment. However, effect estimates were not adjusted for baseline patient and vessel characteristics. A second retrospective study that compared surgical proximal neo-anastomosis versus balloon angioplasty found similar results with primary patency of 71% in the group treated with surgery versus 41% in the group treated with balloon angioplasty [56]. However, results for secondary patency were of similar magnitude and, again, results were not adjusted for baseline differences between the study groups. There was no information on the number of re-interventions.

Translation of the evidence into statements Several surgical and endovascular interventions are available to help non-maturing AV fistulas reach a stage where they can be used successfully for haemodialysis. Both surgical and endovascular procedures achieve moderate primary patency and rather good secondary patency at one year. The variability in outcome for both categories is large, probably due to differences in the study population, and perhaps also due to differences in expertise of the vascular access team. The trade-offs from aggressive efforts to maximize AV fistula maturation may be prolonged catheter use as creation of an alternative permanent vascular access is delayed. Multiple re-interventions may be taxing for patients and ultimately reduce quality of life in comparison with rapid creation of an alternative access or even permanent catheter use. Many of these questions remain unanswered to date.Also, data are limited to primary and secondary patency at one year, and seldom provide insight into true longevity of the AV access. AV fistulas that require intervention before maturation have shorter secondary patency duration than those that mature without an intervention [57]. The cumulative AV fistula survival is markedly inferior in patients requiring two or more interventions to achieve maturation, as compared to those requiring one or no intervention. In addition, AV fistulas requiring more than one intervention to achieve maturation need more interventions to maintain long-term patency once haemodialysis using that AV fistula is started [55].

Comparative studies between surgical and endovascular interventions are scarce, retrospective and uncontrolled for some of the baseline characteristics that may influence both choice of procedure and outcome. With the data currently at hand, the guideline group felt the available evidence to be insufficient to suggest one approach over another.

It seems reasonable to assume clinical multidisciplinary expertise in the absence of clear guidance may be even more important than it is for other areas. Building and nurturing a team of dedicated vascular access specialists may be what maximises success. It allows team members to gain experience in the various techniques available, and to monitor success as well as complications at a local level. In the absence of clear evidence that favours one intervention over another, or even comparative studies

of 226

assessing the trade-offs and harms associated with interventions to aid the non-maturing fistula, at least having a structured approach may benefit outcomes.

Other guidelines on this topicThe Spanish Clinical Guidelines on Vascular Access for Haemodialysis [58]: We recommend a clinical check-up be performed at 4-6 weeks after creation to definitively detect delay or absence of AV fistula maturation from its creation to this moment and elective treatment be proposed. We recommend confirming the suspected lack of maturation by Doppler ultrasound.We suggest early treatment of the non-matured native AV fistula to favour maturation and to prevent thrombosis and definitive loss.We recommend percutaneous or surgical techniques not be used systematically to promote maturation of native AV fistulas.We suggest surgery as the first treatment option (proximal reanastomosis) in native AV fistulas with maturation failure associated with juxta-anastomotic stenosis. In cases where this is not possible, endovascular treatment (percutaneous angioplasty) should be proposed.We suggest significant accessory veins associated with maturation failure be disconnected by percutaneous ligation, surgical ligation or endovascular embolisation with coils. We suggest endovascular treatment be used in the presence of stenosis and surgical treatment when there is no stenosis as the first option, given the lower complexity and healthcare costs.We recommend angioplasty in cases of non-matured native AV fistulas with proximal venous stenosis.We suggest angioplasty of the arterial stenosis when this is the cause of non-maturation of the AV fistula, in cases in which the vascularisation of the limb is not compromised.

CSN, ESVS, KDIGO, NKF-KDOQI, KHA-CARI, or NICE provide no current recommendation on this topic.

Suggestions for future researchGiven the absence of comparative data, a randomised trial comparing surgical versus radiological endovascular interventions would be informative. Investigators should assess long-term outcomes, and record the timing, type, and incidence of interventions required to achieve both maturation and long-term access patency. Only centres experienced in both options should participate. Careful reporting of adverse events and patient quality of life will be instrumental to inform future guidance in this field.

of 226

12. Chapter 4. Self-administered interventions for AV fistula maturation


We suggest a standardised exercise program involving hand and arm exercises may improve AV fistula maturation in adults with end-stage kidney disease. (2C)

There is insufficient evidence to support specific exercise programmes or physical interventions to promote AV fistula maturation in adults with end-stage kidney disease. (-D)

Advice for clinical practice Involving patients more actively in preparing for haemodialysis may improve self-management

skills, health literacy and thereby well-being.

Rationale

BackgroundSuccessful maturation of a newly created AV fistula is essential to allow its use for adequate haemodialysis. Unfortunately, the newly created AV fistula fails to mature in up to a quarter of cases, leading to additional invasive procedures and reduced quality of life. Interventions that ensure improved maturation would surely be welcomed by all. Simple interventions that can be performed by patients themselves, including hand exercise and self-surveillance, seem attractive, presuming they would cause fewer adverse events than medical or surgical interventions. Past guidance suggested performing isometric hand exercise before and following creation of the AV fistula, because it was thought to increase blood flow and therefore enlarge vein diameter [59]. The evidence base supporting these recommendations was patchy and required updating.


We identified three RCTs involving hand or other exercises to enhance maturation of newly created AV fistulas [60-62], and one RCT assessing a device developed to apply reliable intermittent pneumatic compression to the outflow veins [63].We found no RCTs assessing other forms of patient education, patient behaviour, or self-monitoring.

A first RCT compared simple exercises with opening and closing of fingers versus a structured exercise program, which included squeezing a tennis ball, and exercising with a dumbbell and flex-band with a tourniquet positioned proximally to the AV fistula. In the per protocol analysis of 25 participants in both groups, 17 patients performing the simple exercises and 22 following the structured program had matured fistulas two weeks after creation (p= 0.14) [60]. The study was considered at high risk of bias because subjective outcome definition and inadequate random sequence generation.

A second RCT included 18 participants and compared the use of a handgrip versus a soft ball for hand-squeezing exercises. This study only reported surrogate outcomes, such as the increase in vein diameter before and after exercising. The numbers of matured fistulas overall or within each group were not reported [61]. The trial was considered at high risk of bias because of selective outcome reporting and lack of information on important elements in the design and conduct of the study.

of 226

A third RCT included 72 participants who were randomised after proximal or distal AV fistula creation to either a structured program, which consisted of repeated flexion and extension of the elbow and wrist in addition to opening and closing of the hand, or usual lifestyle without specific exercises [62]. After one month, there were numerically more people with a clinically mature AV fistula in the group that had exercised than in the group that had not (95% versus 81%; p=0.07). Also, numerically more AV fistulas were considered ultrasonographically mature among the people who had exercised than among those who had not (82% versus 74%, p=0.45). Importantly though, results were not statistically significant and maturation definitions were based on clinical and radiological criteria rather than on successful dialysis. In addition, although an attempt was made to adequately randomise the participants, due to sample size restrictions important imbalances existed between the groups that could have biased the results. In particular, there were fewer diabetic patients with peripheral vascular disease in the intervention group.

A final RCT assessed a new device developed to apply reliable intermittent pneumatic compression to the outflow veins [63]. The device consists of a miniaturised control unit attached to a wearable pneumatic cuff which inflates to a pressure of 60 mmHg held for 20 seconds and subsequently deflates to 10 mmHg for 55 seconds before the cycle repeats. Forty-eight participants were randomised to either the real or a sham device one week after successful AV fistula creation and asked to wear the device for six hours a day for four weeks. After one month, there was a statistically significant 20% greater increase in venous diameter 5 cm proximal to the arteriovenous anastomosis for the participants in the experimental group. At 10 cm, the average difference was 7% and no longer statistically significant and at 15 cm, there was no difference between the groups. There was no information on maturation. There were no important adverse events reported in the experimental group.

Translation of the evidence into recommendations We found two RCTs, both comparing different self-administered hand exercises. Neither indicated one intervention to be superior over another, but data were sparse and the studies were at high risk of bias. In addition, we found one RCT comparing a structured exercise program versus no exercises, which provided some evidence that such a program may be beneficial. We found this evidence to be of low certainty, due to risk of selection bias and wide confidence intervals from sample size restrictions. More importantly, outcome measures were of a surrogate nature, using clinical and ultrasonographical criteria-based maturation rather than successful dialysis. One month may be too soon to assess the finality of a maturation process and data might have been different if the AV fistulas had been reassessed two weeks later. The guideline development group felt it would be unlikely that simple exercises, such as hand squeezing, could have many harmful outcomes, provided patients waited until sufficient wound healing had occurred. Indeed, the no-exercise controlled trial did not report any important adverse events. Despite the study limitations, the guideline development group felt there was some indication that a structured exercise program could be useful, and would not represent important resource implications, such that in the absence of important adverse events, they supported the use of such programmes in the postoperative phase of AV fistula creation. There was one trial testing a new pneumatic device, but results were considered preliminary and outcomes surrogate in nature.

Other guidelines on this topicESVS [33]Structured post-operative hand exercise programs should be considered, to increase AV fistula maturation. (IIa-B)

NKF-KDOQI [59]Fistula hand-arm exercise should be performed. (B)

The Spanish Clinical Guidelines on Vascular Access for Haemodialysis [58]

of 226

We suggest that the patient do exercises before and after the creation of native AV fistulas to promote maturation.

CSN, KDIGO, KHA-CARI, or NICE provide no current recommendation on this topic.

Suggestions for future researchGiven the scarcity of evidence on self-administered interventions to promote maturation, randomised trials comparing various exercise programmes versus no exercise with adequate sample sizes and standardised outcomes would be informative for future recommendations in this field.

of 226

13. Chapter 5. Perioperative prophylactic antibiotics for preventing AV access infection


We recommend giving preoperative antibiotic prophylaxis for AV graft insertion in adults with end-stage kidney disease. (1C)

We suggest giving preoperative antibiotic prophylaxis for complex AV access procedures in adults with end-stage kidney disease. (2D)

We suggest not giving preoperative antibiotic prophylaxis for simple AV access procedures in adults with end-stage kidney disease. (2D)

Advice for clinical practice Simple AV access procedures include the creation of a native radio-cephalic or native brachio-

cephalic AV fistula. Complex AV access procedures include those that are not considered simple.

Rationale

BackgroundCreation of an AV access for haemodialysis can be considered a clean surgical procedure. Accordingly, preoperative antibiotics should not be necessary and overuse may induce bacterial resistance. However, patients with ESKD have impaired immunity with an increased risk of infection. In addition, the risk and consequences may be dependent on the type of AV access procedure. If prosthetic material is inserted and becomes infected during surgery, the newly created AV access may be jeopardised. As such, recommendations for antibiotic prophylaxis may be different for AV fistula and AV graft procedures and must balance benefits and harms appropriately.


We identified two RCTs assessing perioperative antibiotics for AV graft insertion [64, 65]. No studies were found which assessed prophylactic antibiotics for AV fistula creation.A first RCT included 38 participants, who were randomised to either cefamandole or placebo before insertion of an AV graft in the radio-cephalic (n=19) or femoro-saphenous (n=19) position [64]. Cefamandole or placebo were given intravenously 30 minutes prior to surgery and six to 12 hours postoperatively. The overall infection rate was 26%. Two of 19 participants treated with antibiotics, and eight of 19 participants given placebo, developed an infection (MD -0.32; p<0.04). The study was considered at unclear risk of bias for not reporting the randomisation procedure. The second RCT included 206 patients who underwent a total of 408 procedures to create a permanent vascular access [65]. People were randomised to either a single intravenous dose of 750 mg of vancomycin approximately six to 12 hours before the vascular access placement procedure (206 procedures), or to no prophylactic antibiotics (202 procedures). Within 30 days, and before using the AV access for chronic dialysis, infection developed twice in the antibiotics group and 12 times in the group that had not received antibiotic prophylaxis (MD -0.05; p<0.01). All 14 infections occurred in upper limb polytetrafluoroethylene (PTFE) grafts. The study was considered at high risk of selection

of 226

bias, as the allocation was based on the hospital record number.

Translation of the evidence into recommendations There are no randomised trial data on perioperative antibiotic prophylaxis for AV fistula creation. The guideline development group felt that in the absence of direct evidence, they should rely on extrapolation of evidence for antibiotic prophylaxis for preventing surgical site infections in general. They drew on an evidence review conducted by the British National Institute for Health and Care Excellence in January 2017 [66]. The review process found evidence supporting antibiotic prophylaxis to patients before clean surgery involving the placement of a prosthesis or implant; this was based predominantly on evidence for a clinically relevant reduction in surgical site infections for this category. There is far less evidence related to clean and simple procedures, a single randomised trial indicating evidence for no effect. Our guideline development group considered the creation of a native fistula to be a ‘clean’ and short surgical procedure, in a non-contaminated area. Hence, they judged antibiotic prophylaxis non-mandatory in this setting.

In cases where prosthetic materials are used, two RCTs provided low certainty evidence for a clinically relevant reduction in surgical site infections. This is in line with the conclusion from the evidence review conducted for the NICE guideline [66]. We found no evidence for preferring one type of antibiotic over another in this setting. The guideline development group judged both first generation cephalosporins as well as vancomycin or teicoplanin could be considered, depending upon local practice and epidemiology of methicillin resistance.

Other guidelines on this topicESVS [33]We recommend broad spectrum antibiotics should be given prior to insertion of an AV graft including prophylaxis for Staphylococcus aureus. (IA) In carriers or in units with a high incidence of methicillin resistant Staphylococcus aureus the administration of a parenteral glycopeptide is recommended. (IB)

The Spanish Clinical Guidelines on Vascular Access for Haemodialysis [58]Due to the risk of infection associated with the prosthetic arteriovenous fistula, we recommend the use of perioperative prophylactic antibiotics.

CSN, GEMAV, KDIGO, NKF-KDOQI, KHA-CARI, or NICE provide no current recommendation on this topic.

Suggestions for future researchAn adequately powered RCT comparing routine administration of antibiotics versus no antibiotic prophylaxis before creation of native AV fistulas would be helpful to resolve the remaining uncertainty.

of 226

14. Chapter 6. Timing of first cannulation


AV fistulasIn adults requiring haemodialysis, we suggest AV fistulas can be cannulated four weeks after creation if they are considered suitable for cannulation on clinical examination. (2C)

In adults requiring haemodialysis, we recommend against cannulating AV fistulas sooner than two weeks after their creation. (1B)

In adults requiring haemodialysis, we suggest against cannulating AV fistulas in between two and four weeks after their creation, unless this can avoid placement of a central venous catheter for haemodialysis. (2C)

AV graftsIn adults requiring haemodialysis, we recommend ‘early cannulation type’ AV grafts can be cannulated as soon as wound healing permits. (1B)

In adults requiring haemodialysis, we suggest against cannulating a ‘standard type’ AV graft sooner than two weeks after insertion, unless this can avoid placement of a central venous catheter for haemodialysis. (2B)

Advice for clinical practice In practice, suitability for cannulation on clinical examination is determined by the presence of a

palpable vein and good thrill. If clinical examination is inconclusive, ultrasound with flow measurement may help in deciding

whether or not to cannulate. Bedside ultrasound guided cannulation may be helpful in avoiding complications and decreasing

the number of failed cannulations. Using single needle dialysis, low dialysis blood flows and smaller needles (17 gauge) may prevent

harm to AV fistulas, which are cannulated early. Wound healing refers to the tissue around the body of the graft, rather than the incision site.

Rationale

BackgroundIt has been standard practice to avoid cannulating an AV fistula during the first six weeks after its creation. For standard polytetrafluoroethylene (PTFE) AV grafts this period has always been two weeks, but new generation grafts have been marketed for their early cannulation properties allowing use as an alternative to central venous catheters for prompt access [67]. On the one hand, cannulating a newly created vascular access too early may result in perforation, haematoma or even destruction of the access site. On the other hand, waiting may cause needless insertion of haemodialysis catheters, and lead to delays in searching for causes of non-maturation or in creating an alternative AV access.We wanted to assess when the different AV access types may reasonably be cannulated successfully; and whether certain easily measured variables indicate a first attempt at cannulation is likely to be successful. These variables included clinical observations and ultrasound-based measurements such as vessel diameter, blood flow, wall thickness and depth of the AV access.

Summary of the evidence(Supplement 3| Review Questions – PICOM Format - Chapter 6)(Supplement 4| Search Strategies - Chapter 6)

of 226

(Supplement 5| Study selection flow diagrams - Chapter 6)(Supplement 6| Summary evidence tables - Chapter 6)

AV fistulasWe found no RCTs assessing the effect of timing of first cannulation on outcome in AV fistulas.

Eight observational studies assessed the effect of the time point of first cannulation after AV fistula creation on outcome [68-75]. Results were conflicting. Two studies based on the Dialysis Outcomes Practice Patterns Study (DOPPS) provided centre-level information on timing of first cannulation [68, 69]. When analysing all patients who started haemodialysis with an AV fistula (n=894), median time to first cannulation varied greatly between countries: averaging 25 days for Japan and 27 days for Italy, 42 days for Germany, 80 and 86 days for Spain and France respectively, and 96 and 98 days for the UK and US [68]. No association was found between cannulating sooner versus later than 28 days after AV fistula creation on the one hand, and age, sex, and fifteen different classes of variable reflecting comorbidity on the other hand. Cannulating an AV fistula sooner than 14 days after its creation was associated with a twofold increased risk of subsequent fistula failure versus cannulating after 14 days (p<0.01). There was no difference for AV fistulas cannulated within 15 to 28 days compared with fistulas cannulated within 43 to 84 days. Based on the same cohort, investigators analysed the influence of time to first cannulation and blood flow rate at first cannulation on primary fistula failure [69]. For newly created AV fistulas, a first attempt at cannulation occurred within two months after placement in 36% of American, 79% of European and 98% of Japanese facilities. There was no indication that centres which cannulated their AV fistulas within four weeks after creation had higher risks of later fistula failure. Facility median blood flow rate was not significantly associated with access failure either. A retrospective study, including 190 Moroccan haemodialysis patients, found no difference in the risk of fistula thrombosis whether cannulation occurred before or after 21 days following AV fistula creation [70]. However, the authors did not provide numeric data, there was no attempt at adjusting for confounders and it was unclear when or how outcomes were measured. A prospective cohort study enrolled 118 people with a newly created AV fistula [71]. In contrast to the DOPPS data, postponing cannulation until one month after fistula creation reduced the risk of thrombosis by 60% (RR 0.40 – 95%CI 0.19 to 0.84). This finding was supported by a second prospective cohort including 535 people with newly created AV fistulas [72]. In this study, which adequately controlled for possible confounders, cannulating AV fistulas within 30 days of their creation was associated with twice the hazard for primary AV fistula failure. Two studies published by a group from the UK retrospectively reviewed fistula patency in 1167 fistulas created between 2002 and 2015 with different cut-off points to define early cannulation [73, 74]. Both analyses were unadjusted comparisons using the AV fistula as the unit of analysis rather than the individual patient. The first report used a cut-off of 30 days after fistula creation to define early and late cannulation and found no difference in fistula failure within 90 days following first cannulation between the two groups (p=0.35) [73]. The second report separately analysed AV fistula survival in pre-dialysis patients and patients on maintenance haemodialysis with a definition of early cannulation as an attempt that happened within four weeks after fistula creation. It found no difference in fistula survival between the cohorts [74]. One study prospectively included all adults with chronic kidney disease set to undergo upper extremity AV fistula creation at one of seven American centres to assess the association between a variety of clinical processes, cannulation and successful haemodialysis [75]. In a subset already treated with chronic haemodialysis at the time of AV fistula creation, the investigators analysed the influence of timing to first cannulation on overall maturation, broadly defined as successful dialysis during a four-week period. They found that for every month a first cannulation attempt was delayed beyond two months, the odds for successful dialysis gradually diminished (OR 0.93; 95%CI 0.89 to 0.98). The certainty of the evidence was affected by possible residual confounding and inflated type-1 error from multiple analyses.

Four studies assessed the value of clinical assessment or ultrasound-based measures in determining readiness for first cannulation [76-79].

of 226

A first study assessed whether ultrasound and clinical evaluation of AV fistulas at four weeks after creation predicted future successful cannulation [76]. Clinical appreciation, by experienced nurses unaware of the ultrasound findings, consisted of examining the flow characteristics (categorised as thrill/pulse/audible bruit with or without a palpable thrill), the vein calibre, and the straightness and the depth of the vein. Clinical appreciation, a vein diameter >5 mm, and arterial end diastolic velocity >110 cm/sec had a positive predictive value for successful cannulation of 81%, 90% and 87%, respectively. The negative predictive value, however, was only 63%, 53% and 37%, respectively. Importantly, this study excluded patients with AV fistulas that had already thrombosed prior to or at four weeks. A second, single centre consecutive cohort of 20 patients, assessed whether venous wall thickness and circumferential stress could predict successful cannulation, defined as successful dialysis without substantial extravasation of blood at the cannulation site [77]. The investigators used very high frequency-based assessment (55 MHz probe) of intimal media thickness, defined as the sum of a thin blood-intimal interface and a uniform hypo-echogenic media. A minimum threshold intimal media thickness of 0.13 mm was associated with successful cannulation without extravasation (sensitivity 87%, specificity 92%). Importantly, patients were only cannulated if the venous diameter was >6 mm, the blood flow >600 mL/min, and the depth of the fistula <6 mm for a segment >6 cm six weeks after AV fistula creation. A third cohort study retrospectively assessed the maturation surveillance ultrasound examinations of 69 patients, performed for assessing fistula maturation within four months of AV fistula creation [78]. A venous diameter >0.4 cm or a blood flow >500 mL/min were reasonable predictors of successful cannulation within four months. Whereas in patients who met both criteria, cannulation was successful in 19/20, in those who failed both criteria, successful cannulation occurred in only 5/15. In this study, clinical assessment by experienced nurses correctly predicted successful cannulation in 24/30 patients (accuracy of 80%). Finally, a fourth study recorded fistula flow and diameter for 12 weeks following creation of the AV fistula. None of 14 fistulas with a diameter increase <0.4 cm 14 days after creation matured, while all 38 fistulas with a diameter increase >0.4 cm were successfully cannulated within 12 weeks after creation. The same results were seen with a cut-off blood flow of less or more than 400 ml/min [79].

AV GraftsWe found one RCT comparing cannulation within one to two days versus after 10 to 14 days after insertion of an upper limb standard PTFE AV graft in 36 patients requiring semi-urgent dialysis. There were no meaningful differences in the number of participants with haematomas, thrombosis or infection. There was no meaningful difference in compression time to control bleeding or venous back pressure. Evidence was considered moderate to low certainty because of unclear risk of bias and wide confidence intervals [80].

We found an additional six observational studies covering AV grafts [69, 71, 81-84]. All directly assessed the effect of the time point of first cannulation after AV graft creation on outcome; four in prospective [69, 71, 83, 84] and two in retrospective cohort studies [81, 82].Two prospective cohort studies reported on cannulation within the first three days after creation [83, 84]. The first study assessed 76 stretch-expanded PTFE grafts and found no meaningful difference in primary patency after three and 12 months for those cannulated after one to two days, versus those cannulated after two weeks [84]. A second study from the US reported an unadjusted comparison of early cannulation within 72 hours of graft placement to late cannulation >21 days following graft placement in 87 patients, using a multilayer graft designed for early cannulation. They found no meaningful difference in cumulative graft patency rates at up to 12 months of follow-up (76% versus 77.5%; p=0.7) [83].

Three studies reported on cannulation of standard AV grafts within 14 days and compared the results with AV grafts cannulated after 14 days [69, 81, 82].In the DOPPS, first cannulation of AV grafts typically occurred within two to four weeks at 62% of American, 61% of European and 42% of Japanese facilities [69]. In 17% of American, 16% of European and 42% of Japanese facilities centres, which cannulate their AV graft within two weeks of

of 226

creation, overall risk of AV graft failure, defined as the time to first thrombosis or access salvage procedure did not differ from those cannulating at later times. A retrospective study reported on the 12-month failure of 64 AV grafts in 58 people who had their standard AV graft’s first cannulation at different times after insertion, starting from the second week after surgery. The study results suggested that timing of the first cannulation of a standard AV graft had no significant impact on graft survival [81]. The overall incidence of primary AV graft failure, defined as the first occurrence of graft thrombosis or need of any invasive access procedure, and of cumulative graft failure, defined as irreparable loss of the graft, at 12 months was 72% and 41% respectively. There was no association between the time of first cannulation and cumulative graft loss. Primary graft failure seemed to decrease as the interval between the procedure and first cannulation increased, but results were not statistically significant and certainty of the evidence was very low due to the requirement of successful first cannulation as an inclusion criterion and censoring for patient mortality (15% grafts), disproportionally present in the later time groups. A third and retrospective study in 270 people receiving a standard AV graft found secondary patency rates up to 15% lower in grafts cannulated within versus after 14 days. Certainty of the evidence was very low, as we considered the study at high risk of bias with comparisons of raw numbers, not adjusted for confounders [82].Finally, a prospective cohort in North America enrolling 147 people with AV grafts found first cannulation after versus within one month had uncertain effects on risk of thrombosis within one year of cannulation (n=147; RR 0.77 – 95%CI 0.43 to 1.38). The analysis had been adequately adjusted for the main confounding factors [71].

Translation of the evidence into recommendations Several observational studies consistently indicate that cannulating an AV fistula within 14 days of its creation importantly increases – almost doubles - the risk of unsuccessful dialysis and/or later AV fistula failure compared with cannulating an AV fistula after 14 days. The evidence for waiting another 14 days is less impressive and not consistent. In addition, the negative effects of a further delay, i.e. the need for urgent central venous catheter placement, have never been never studied and may counterbalance the positive effects of fistula longevity. In the absence of this evidence, the guideline development group felt that in this case avoiding placement of a catheter weighed more heavily and allowing another 14 days for further maturation weighed less, in comparison with the previous case. In the absence of the need for urgent dialysis, it seems reasonable to allow for an additional 14 days of further maturation before attempting to cannulate the AV fistula. This also holds true for those already receiving dialysis via a tunnelled catheter, unless a problem with the catheter should arise.

AV fistulas with a palpable vein and good thrill at four weeks after their creation, can be cannulated successfully in most cases. In this situation, additional ultrasound measures are unlikely to be helpful. However, in the absence of such a thrill, there is low quality evidence in line with clinical practice suggesting that an AV fistula diameter of >4-5 mm or a blood flow >500 mL/min indicate the fistula has matured and can be cannulated successfully. In the absence of a thrill a diameter of <4 mm and a blood flow of <400 ml/min make it highly suspicious that the AV fistula will fail without intervention. Although other techniques to assess AV fistula characteristics have been proposed, further study is needed to assess their added value.

One small RCT and several observational studies provide moderate certainty evidence that cannulating an AV graft within two days of its insertion has no negative consequences for short- or long-term AV graft outcome, including infection rates. This is the case even with standard PTFE grafts. There does not seem to be an increase in the complication rate but early cannulation of standard PTFE grafts has never found its way into routine practice around the world. Randomised controlled trails of the new grafts designed for early cannulation are not available. One retrospective study showed no increase in complications when cannulation of an early cannulation graft within the first 72 hours was compared with cannulation after 3 weeks. How this influences the added benefit of avoiding temporary and tunnelled central venous catheter placement is unclear, but it can only be expected to further tip the benefit -harm balance in favour of supporting early cannulation when necessary.

of 226

Other guidelines on this topicESVS [33]AV fistulas should be considered for cannulation 4 – 6 weeks after creation, and standard AV grafts after 2 – 4 weeks. (IIa-B). AV fistula cannulation before 2 weeks should generally not be done. (III-C) AV fistula cannulation between 2 and 4 weeks after creation may be considered in selected patients under close supervision. (IIIb-B)

The Spanish Clinical Guidelines on Vascular Access for Haemodialysis [58]We recommend that cannulation of the native AV fistula not be initiated in the first two weeks following creation and that the optimal time for the first cannulation be decided on a case-by-case basis.We recommend that cannulation of the prosthetic AV fistula be initiated between 2 and 4 weeks following construction, except in those of immediate cannulation.

CSN, KDIGO, NKF-KDOQI, KHA-CARI, or NICE provide no current recommendation on this topic.

Suggestions for future researchGiven the lack of high-quality comparative data, randomised trials comparing decision rules to determine optimal time points for first cannulation in grafts and fistulas would be informative. These trials should take care to investigate objective, and reproducible criteria for deciding upon timing of first cannulation of a vascular access. Long-term access outcomes, adverse events, and quality of life measures should be reported transparently.

of 226

15. Chapter 7. Vascular access surveillance


AV fistulasWe suggest the evidence for technical surveillance in addition to clinical monitoring of a functional AV fistula to detect and pre-emptively correct a haemodynamically important AV access stenosis in adults is inconclusive and needs more research. (2C)

AV graftsWe suggest against technical surveillance in addition to clinical monitoring of a functional AV graft to detect and pre-emptively correct a haemodynamically important AV access stenosis in adults, unless it occurs in the context of a clinical study. (2C)


Rationale

BackgroundAV fistulas or grafts can develop stenotic lesions anywhere in the arterio-venous circuit due to neointimal hyperplasia. This may lead to dysfunction or thrombosis of the vascular access, making it unfit for use. It happens quite often and is associated with an increased risk of irremediable access failure [85]. Clinical monitoring is defined as clinical assessment of an AV access at regular intervals including examination of the AV access thrill and bruit, haemostasis time after needle removal and outflow appraisal after arm elevation. Technical surveillance is defined as the assessment of an AV access at regular intervals using specialised apparatus. Both clinical monitoring and technical surveillance have been advocated under the assumption that detection of a haemodynamically important AV access stenosis (>50% reduction in luminal diameter), and subsequent intervention to prevent further vessel narrowing and thrombosis, leads to improved longevity of the AV access compared with salvage interventions deferred to when the access becomes dysfunctional [2, 86]. However, such a practice may inadvertently lead to more invasive diagnostic and therapeutic interventions and expose patients to the complications thereof.


A Cochrane systematic review, with content assessed as up-to-date to 30 November 2015 included 30 reports of 14 RCTs comprising 1390 participants [87]. Nine studies enrolled adults without a documented or suspected access stenosis (primary prophylaxis), five enrolled adults with either a documented or suspected stenosis in a functioning access (secondary prophylaxis). In primary prophylaxis, pre-emptive correction followed the identification of a stenosis using various technical surveillance strategies: Doppler ultrasound, sequential access blood flow measurements, or measurements of dialyser outlet pressure; often in addition to clinical monitoring. In secondary prophylaxis, participants were randomised either before or after angiographic confirmation of a stenosis to pre-emptive correction or continued monitoring until the access became dysfunctional. Five studies included only AV fistulas, eight only AV grafts, and one included both. Follow-up ranged

of 226

from six months to three years. Attaining a maximum 11/11 on the AMSTAR checklist, we considered the review to be of high quality and its results trustworthy. In what follows, we summarise the findings of that review. A meta-analysis including data from five randomised trials, indicated that pre-emptive correction had uncertain effects on patient death. The point estimate favoured no intervention, but the confidence interval spanned a potentially important reduction and increase in risk, such that the overall level of evidence for effect was very low [87].Pre-emptive correction may have slightly reduced permanent access loss. Both because of high risk of bias in the included studies and the width of the confidence interval, the certainty of the evidence was low. The authors of the review decided to subgroup studies according to whether participants had an AV fistula or an AV graft and concluded that pre-emptive correction may reduce the risk of permanent access failure in AV fistulas but may make little or no difference in AV grafts. From visual inspection of the forest plots, we understand the decision to conduct the subgroup analysis, although there was no statistical indication of heterogeneity in the primary analysis. Hence the evidence supporting a claim of differential effect remains low. Similarly, the effect does not seem to have depended on whether studies covered primary or secondary prophylaxis, on the type of surveillance strategy used, or the type of remedial procedure performed.

Twelve RCTs assessed the effect of pre-emptive correction on access thrombosis (possibly remediable access failure). Overall, pre-emptive correction slightly reduced the risk of access thrombosis, but there was a moderate degree of heterogeneity in the analysis, which could be explained by the modifying effect of access type. In subgroup analysis there was moderate certainty evidence for an important reduction in the risk of possibly remediable failure of an AV fistula. Conversely, there seemed to be little or no effect on the risk of possibly remediable failure of an AV graft. Another source of heterogeneity involved the aim of prevention. In subgroup analysis, the data indicated that the effect may be different for people in whom a stenosis was already suspected or documented (secondary prophylaxis) from those in whom that was not the case (primary prophylaxis), but a lot of heterogeneity remained in the analyses. In primary prophylaxis, the type of surveillance strategy had no visual or statistical influence on the effect of pre-emptive strategies.

For infections, the evidence was very limited. Only three RCTs included in the review assessed this outcome, but all three defined the outcome differently: one did not provide a definition, one included only access infections, and one only catheter-related infections. Given that pre-emptive correction may decrease catheter use (see below), effects on catheter-related infection and AV access infection are expected and indeed appeared to be in opposite directions, invalidating meta-analysis of these studies in our view. At this point in time, it remains unclear how the intervention affects net infection risk as no study assessed both outcomes in the same study.

Pre-emptive correction probably increases the number of diagnostic angiograms in people that receive primary surveillance of their AV access, and undoubtedly in those who already have a suspected stenosis based on clinical monitoring or other technical surveillance strategies (moderate certainty evidence). It may decrease catheter use, and risk of hospital admission, but the level of evidence remains low due to risk of bias in the primary studies and a large amount of unexplained heterogeneity in the analysis with point estimates on opposite sides of the line of no effect.

In addition to the Cochrane review, we identified two reports of one more recent randomised trial, which had not yet been included in the systematic review [88, 89]. The study enrolled 212 adults who had dialysed successfully via an AV fistula for the past three months. Participants were randomised to a ‘classic’ or ‘access blood flow based’ surveillance program. The classic surveillance program included clinical monitoring, i.e. physical examination of the AV fistula before every dialysis; and ‘first generation’ or ‘classic’ technical surveillance through effective blood flow, dynamic dialyser inlet and outlet pressure measurement during every dialysis; weekly Kt/V assessment; and recirculation measurement using the urea method every three months. Presence of any of seven alarm criteria would trigger fistulography and subsequent percutaneous transluminal angioplasty or surgical intervention depending on the findings. The access blood flow-based surveillance program included

of 226

Doppler ultrasound and Doppler dilution methods to assess blood flow in addition to what was used in the classic surveillance program, on a three-monthly basis. Classic criteria would trigger additional access blood flow measurement. A set of three criteria (blood flow decrease >25%; blood flow <500 mL/min; >50% reduction vessel diameter plus either peak systolic blood velocity >400 cm/sec or ratio pre- and post-stenosis >3) would trigger fistulography and subsequent treatment if necessary. At the end of a three-year follow-up – with staggered censoring – they found an important increase in thrombosis-free patency (HR 0.38 – 95%CI 0.11 to 0.82) and access survival until abandonment (HR 0.49 – 95%CI 0.26 to 0.93) for those receiving access blood flow-based surveillance. Intervention-free access survival was similar (HR 0.98 – 95%CI 0.57 to 1.61). Also, the number of interventions during the entire follow-up was not different (0.14/patient/year in both groups).

Translation of the evidence into recommendations For a screening programme to be successful, two important elements are needed. Not only should the screening test be effective at detecting the presence of an underlying significant stenosis, there should also be evidence that subsequent correction of the stenosis prolongs AV access survival.

In weighing benefits against harms, the guideline development group assigned the most value to patient survival and permanent access loss. The evidence to date indicates that technical surveillance and subsequent pre-emptive correction of an AV access stenosis may possibly and slightly reduce the risk of permanently losing an AV fistula. It also appears that this effect may be smaller for AV grafts, if it exists at all. This is regardless of which surveillance technique is used or which intervention subsequently performed. In addition, there is moderate quality evidence that even possibly remediable access failure is probably not importantly reduced by pre-emptive intervention, whatever the intervention may be.

For AV fistulas, technical surveillance and pre-emptive correction seems to have a larger effect than the overall estimate indicated, but caution is required in interpreting both the relative and absolute effect sizes obtained by the review. First, although visual inspection of the forest plot indicated effect modification by access type, there was no statistical indication that heterogeneity truly exists. Translating the obtained subgroup effect estimate may thus overestimate the true effect. A more conservative estimate assumes the overall relative risk of 0.8 with its confidence interval. The corresponding absolute effect heavily depends on the baseline risk of access failure in the control group, which is expected to be (much) larger in the people already suspected to have an access stenosis than in those who are not. Estimating baseline risk from the studies the relative effect of 0.8 translates into an estimated 5 fewer AV fistulas being lost for every 100 patients screened and an estimated 6 fewer for every 100 patients undergoing pre-emptive correction of a documented stenosis after one year. There is better quality evidence for AV fistula thrombosis. There is moderate quality evidence that surveillance and pre-emptive correction moderately reduces the risk of fistula thrombosis, the relative risk of 0.5 translating into an estimated absolute 15 fewer AV fistula thromboses for every 100 patients surveilled for one year and an estimated 23 for every 100 patients undergoing pre-emptive correction of a documented stenosis. This needs to be weighed against the increased number of diagnostic angiograms, which may ultimately not change the number of invasive procedures, a person needs to undergo. The value patients put on being able to have these planned – in case of surveillance, rather than having to undergo them in emergency setting – in the case of access thrombosis, may sway the balance of perceived benefits and harms. Fewer catheters may be required, but the overall effect on infection rate remains unclear to date. Additional demands on individual radiology services may also limit the feasibility of routine surveillance programmes. Because of uncertainties around the absolute reduction in risk of AV fistula failure, which needs to be weighed against an increased number of diagnostic angiograms, the guideline development group ultimately refrained from speaking out for or against technical surveillance. A more recent RCT compared two strategies of surveillance: ‘classic’ or first-generation versus ‘classic plus access blood flow-based’ or second generation surveillance [88]. There was moderate evidence that access blood flow-based surveillance resulted in reduced access thrombosis, and reduced AV fistula abandonment without increasing the total number of interventions patients had to undergo.

of 226

Although this does not directly answer the question, it seems to indicate the superiority of access blood flow-based surveillance over classic surveillance methods. However, the GDG felt that at this stage, more research was needed before any specific recommendation could be made.

Other guidelines on this topicESVS [33]Routine physical examination is recommended for vascular access surveillance and maintenance. (I-B)It is recommended that vascular access monitoring is performed by flow measurement of AV grafts monthly and AV fistulas every 3 months. (I-B)When AV fistula blood flow measurement during dialysis indicates the presence of a vascular access stenosis, Qa is <500 mL/min, angiographic assessment of stenosis should be considered. (IIa-B)Venous pressure adjusted for the mean arterial pressure >.50 (or derived static venous pressure >.55) is not a reliable indicator of stenosis and intervention based on this finding is not recommended. (III-C)When haemodialysis efficiency is impaired, investigation and correction of an underlying vascular access stenosis should be considered. (IIa-B)Surveillance of AV fistulas with duplex ultrasound at regular intervals and pre-emptive balloon angioplasty should be considered to reduce the risk of AV fistula thrombosis. (IIa-A)Surveillance of AV grafts with duplex ultrasound at regular intervals and pre-emptive balloon angioplasty is not recommended to prevent thrombosis or improve AV graft functionality. (III-A)

CSN [90] Measure access flow bimonthly in AV fistulas and venous pressure or access flow monthly in AV grafts. (Grade D)

Perform angiography if fistula flow decreases to <500 ml/min or drops >20% from baseline (Grade D); if AV graft flow decreases to <650 ml/min or drops >20% from baseline. (Grade D)

NKF-KDOQI [86]Prospective surveillance of fistulas and grafts for haemodynamically significant stenosis, when combined with correction of the anatomic stenosis, may improve patency rates and may decrease the incidence of thrombosis.

The Work Group recommends an organized monitoring/surveillance approach with regular assessment of clinical parameters of the AV access and HD adequacy. Data from the clinical assessment and HD adequacy measurements should be collected and maintained for each patient's access and made available to all staff. The data should be tabulated and tracked within each HD centre as part of a quality assurance program.

Physical examination should be used to detect dysfunction in fistulas and grafts at least monthly by a qualified individual. (B)

Techniques, not mutually exclusive, that may be used in surveillance for stenosis in grafts include: Preferred:

o Intra-access flow by using 1 of several methods that are outlined in Table 7 using sequential measurements with trend analysis. (A)

o Directly measured or derived static venous dialysis pressure by 1 of several methods. (A) (Protocol provided in Table 8 for using transducers on HD machines to measure directly; criteria in Table 9 for derived methods.)

o Duplex ultrasound. (A) Acceptable:

o Physical findings of persistent swelling of the arm, presence of collateral veins, prolonged bleeding after needle withdrawal, or altered characteristics of pulse or thrill in a graft. (B)

of 226

http://kidneyfoundation.cachefly.net/professionals/KDOQI/guideline_upHD_PD_VA/va_guide4.htm#table9



Unacceptable:o Unstandardised dynamic venous pressures (DVPs) should not be used.

Techniques, not mutually exclusive, that may be used in surveillance for stenosis in AV fistulas include:

Preferred:o Direct flow measurements. (A)o Physical findings of persistent swelling of the arm, presence of collateral veins,

prolonged bleeding after needle withdrawal, or altered characteristics of pulse or thrill in the outflow vein. (B)

o Duplex ultrasound. (A) Acceptable:

o Recirculation using a non–urea-based dilutional method. (B)o Static pressures (B) direct or derived. (B)

One should not respond to a single isolated abnormal value. With all techniques, prospective trend analysis of the test parameter has greater power to detect dysfunction than isolated values alone. (A)

Persistent abnormalities in any of the monitoring or surveillance parameters should prompt referral for access imaging. (A)

An access flow rate less than 600 mL/min in grafts and less than 400 to 500 mL/min in fistulas. (A)

A venous segment static pressure (mean pressures) ratio greater than 0.5 in grafts or fistulas. (A)

An arterial segment static pressure ratio greater than 0.75 in grafts. (A)

NICE [91]Quality standard: Adults receiving haemodialysis have their vascular access monitored and maintained using systematic assessment.

The Spanish Clinical Guidelines on Vascular Access for Haemodialysis [58] We recommend performing a complete physical examination of the AV access in every advanced chronic kidney disease clinic visit, to assess maturation and to detect early on any complication before the first cannulation.We recommend that Doppler ultrasound be performed if insufficient development of a native arteriovenous fistula is observed during physical examination in regular advanced chronic kidney disease outpatient check-ups.We recommend that haemodialysis units have protocolised programmes for AV fistula follow-up, involving multidisciplinary participation. These programmes should include methods for early diagnosis of AV fistula dysfunction and locate its origin, as well as performing the elective treatment.We recommend that the application of programmes for AV fistula follow-up must involve periodic assessment of the parameters obtained by each monitoring and/or surveillance method applied.We recommend that the repeated alteration of any monitoring and/or surveillance parameter be used as a criterion to perform an imaging examination of the AV fistula in front of suspected pathology.We recommend that both Doppler ultrasound and dilution screening methods be used interchangeably to assess AV fistula function, as they have an equivalent performance for blood flow determination.We recommend that Doppler ultrasound be used as the first-choice imaging test in the hands of an experienced examiner, without the need for confirmatory fistulography, to indicate elective treatment in the event of suspected significant stenosis. We recommend that fistulography be reserved as a diagnostic imaging exploration only for cases with inconclusive Doppler ultrasound findings and persistent suspicion of significant stenosis.

of 226

According to the current concept of significant stenosis, we do not recommend that surveillance of the prosthetic AV fistula be performed using second-generation screening methods, whether they be dilution methods to estimate the blood flow or Doppler ultrasound.According to the current concept of significant stenosis, we recommend that first-generation screening methods be used for monitoring the prosthetic AV fistula.According to the current concept of significant stenosis, we recommend that both first- and second-generation methods be used for monitoring and surveillance of the native AV fistula.We recommend that a stenosis be considered significant when there is any reduction in the vascular lumen in native or prosthetic AV fistulas, shown by Doppler ultrasound, which meets all the criteria for high risk of thrombosis (the 2 main criteria and at least 1 additional criterion).We recommend that an elective intervention be performed without delay by percutaneous transluminal angioplasty and/or surgery when the diagnosis of significant AV fistula stenosis is established because of the high risk of thrombosis.We recommend that a stenosis be considered non-significant when there is any reduction in the vascular lumen in native and prosthetic AV fistulas, shown by Doppler ultrasound, which does not meet all the criteria for high risk of thrombosis.We recommend that the elective intervention not be performed when a diagnosis of non-significant stenosis is established in an AV access because of the low risk of thrombosis.We recommend that all non-significant AV fistula stenosis be strictly controlled using second-generation screening methods due to the risk of progression to significant.We recommend elective intervention be performed on the dysfunctional AV fistula with significant stenosis instead of restoring after thrombosis.We suggest surgical treatment of juxta-anastomotic stenosis of the native arteriovenous fistula be performed provided a central venous catheter does not need to be placed.We suggest venous juxta-anastomotic stenosis of the prosthetic arteriovenous fistula be treated indistinctly by angioplasty or surgical intervention.We suggest non juxta-anastomotic stenosis of the native AV fistula initially be treated using angioplasty because it is less invasive than surgery.We recommend fistulography be performed if central venous stenosis is clinically suspected.We recommend only central vein stenosis that are symptomatic be treated.We recommend endovascular therapy be performed using percutaneous transluminal angiography with balloon as the first treatment option for central stenosis.We suggest the use of stents be limited to selected cases where there is technical failure of angioplasty and frequent relapse of stenosis, and we recommend they not be used in venous confluents.We suggest that angioplasty be used as the initial treatment in stenosis in the cephalic vein arch. Treatment by stent placement or by surgical transposition of the cephalic vein may also be considered.

UK Renal Association [92]We recommend that all patients on long term haemodialysis should have their vascular access monitored and maintained to minimise failure, to allow timely planning for subsequent replacement with definitive vascular (or peritoneal) access and to avoid the need for emergency access. (1B)We suggest that systematic observation and advanced surveillance should be employed to predict and prevent access failure. (1C)

KDIGO or KHA-CARI provide no current recommendation on this topic.

Suggestions for future researchGiven the possible, but insufficiently clear, benefits of screening and subsequent pre-emptive correction of an established stenosis in functioning AV fistulas, an adequately powered RCT assessing the right outcomes (AV fistula loss, death, quality of life, infections) would be informative. As screening can only ever be useful if interventions for correcting established AV fistula stenosis are effective, it would be pragmatic to enrol only participants with a suspected or documented access stenosis. An informative RCT would include meticulous records of how many patients were screened, how often, what monitoring, and surveillance methods were used, and how much staff time was spent on access screening, data collection and interpretation. Ravani and co-workers estimated that based on

of 226

the findings of their review, an RCT of 1020 participants per arm, recruited over one year and followed for three years would have a power of 90% to detect a 30% reduction in the hazard ratio for access loss as significant at a two-sided p-value of 0.01, assuming a baseline risk of 10% and a withdrawal rate of 10% [87]. To better inform best practice recommendations in this field, patient preferences concerning the various outcomes and their views on elective versus urgent intervention should be included in future research.

of 226

16. Chapter 8. Medical treatments for maintaining long term AV access patency


AV fistulasWe suggest any decision to give fish oil to adults with end-stage kidney disease in the year after AV fistula creation, must balance improved patency at one year against an unknown risk of bleeding and other side effects. (2C)

We suggest far infrared therapy may be considered for improving long term AV fistula patency in adults with end-stage kidney disease. (2C)

There are insufficient RCT data to make a recommendation for aspirin, clopidogrel, ticlopidine, warfarin, sulphinpyrazone, vonapanitase, beraprost sodium, cholecalciferol, statins, dipyridamole or dipyridamole combined with aspirin to be given for maintaining long-term AV fistula patency in adults with end-stage kidney disease. (-D)

AV graftsWe recommend against warfarin in combination with antiplatelet agents, and against clopidogrel in combination with high dose aspirin for reducing AV graft thrombosis in adults with end-stage kidney disease. (1C)

We suggest any decision to give fish oil in the year following AV graft creation in adults with end-stage kidney disease must balance any improvement in graft patency at one year against an unknown risk of bleeding. (2C)

There are insufficient RCT data to make a recommendation for aspirin, clopidogrel, ticlopidine, warfarin, beraprost sodium, statins, dipyridamole or dipyridamole combined with aspirin to be given for maintaining long-term AV graft patency in adults with end-stage kidney disease. (-D)

Advice for clinical practice

-

Rationale

BackgroundAV fistulas or grafts can develop stenotic lesions anywhere in the arterio-venous circuit, mostly due to neointimal hyperplasia. This may lead to dysfunction or thrombosis of the vascular access, making it unfit for use. AV access thrombosis happens quite often and is associated with an increased risk of irremediable access failure [85]. Several medications, including antiplatelet agents and vitamin K antagonists, are thought to prevent AV access thrombosis and increase AV access patency and longevity, but may also cause bleeding [93]. As immediate maturation problems in the first weeks and months after access creation may result from different pathophysiological factors than patency problems in the long run, the two issues are discussed in two different sections. In this chapter, we assess long-term patency, maturation is discussed in chapter 1.

Summary of the evidence(Supplement 3| Review Questions – PICOM Format - Chapter 8)

of 226

(Supplement 4| Search Strategies - Chapter 8)(Supplement 5| Study selection flow diagrams - Chapter 8)(Supplement 6| Summary evidence tables - Chapter 8)

Five systematic reviews of RCTs assessing benefits and harms of various medical adjuvant treatments to increase patency of AV fistulas and AV grafts were identified. We judged all these reviews to be of moderate to high quality with AMSTAR scores of 8 to 10/11 [13-17]. All the reviews included both studies measuring patency outcomes after six weeks to 12 weeks, as well as patency outcomes measured several months later. Based on group consensus, for this section, we chose to only consider studies measuring patency outcomes after 12 weeks, as an arbitrary cut-off to distinguish maturation from long-term patency.

The first was a Cochrane systematic review with content assessed as up-to-date to 23 March 2015 [13]. It included 15 RCTs comprising 2230 participants. Seven trials included people with an AV fistula, six with an AV graft, and two with either an AV fistula or AV graft. After exclusion of the studies on maturation (outcomes registered within 12 weeks of creation), nine studies remained for consideration in the present section. Three of these included AV fistulas; the remaining six included AV grafts. All but one enrolled participants at the time of AV access insertion [19]. Tested medical interventions included aspirin, dipyridamole, dipyridamole plus aspirin, warfarin, fish oil, clopidogrel, and human type I pancreatic elastase. Studies mostly included participants of all ages, follow-up ranged from five to 18 months after AV access insertion. Most studies only assessed AV access thrombosis as primary outcome of interest.

The second was also a Cochrane systematic review, which covered specifically anti-platelet agents to prevent vascular access failure, and other outcomes in people with chronic kidney disease, with content assessed as up-to-date to 24 January 2011 [14]. Most of the data contained in that review were not included in this section, as outcomes were mostly registered within three months of AV access creation. This systematic review focused only on the effect of antiplatelet agents, and the included studies largely overlapped with the ones that Tanner and co-workers later included in their review [13]. But in contrast to that review, the second review also included existing access systems [14], which resulted in the identification of two additional RCTs for this guideline section [94, 95].

Three other systematic reviews each assessed a specific treatment: fish oil [15] and far infrared therapy [16, 17]. In addition to these reviews, a further three RCTs, published after 2013, and not included in any of the included systematic reviews, assessing various adjuvant medical treatments for improving patency were identified [23, 24, 96].

AspirinTwo RCTs compared different doses of aspirin versus placebo in AV grafts with overall very uncertain effect on AV access thrombosis [19, 97]. The included two studies were small and at unclear risk of bias, results were very inconsistent, and the confidence interval was very wide spanning the line of no effect.A third RCT compared 30 mg of aspirin daily versus placebo in a random cross-over design in 137 patients, who had been on haemodialysis for more than six weeks, while treated with erythropoietin [95]. The investigators found no appreciable difference in thrombosis or loss of patency. They reported no data on adverse outcomes.

ClopidogrelA first RCT assessed the combination of clopidogrel and aspirin versus placebo for AV graft outcomes in newly inserted AV grafts [94]. The study was terminated early after 12 months because of excessive bleeding risk in the active treatment arm. Antiplatelet treatment did not alter thrombosis or loss of patency.A second study randomised 96 participants to receive either clopidogrel plus a prostacyclin analogue or placebo for 7 days before until one year after AV fistula creation [23]. The investigators described

of 226

an important reduction in primary fistula failure at one year. They reported no bleeding, but external validity of the results was considered low because of stringent exclusion criteria. Only 25% of all people set to undergo AV fistula creation had been enrolled.

DipyridamoleOne RCT tested the effect of dipyridamole or dipyridamole plus aspirin versus placebo on thrombosis within 18 months in people with newly inserted AV grafts [97]. The point estimate favoured dipyridamole in both cases, but certainty of the evidence was low due to an unclear risk of selection and selective reporting bias, as well as a very wide confidence interval spanning both an important reduction and increase in graft thrombosis. Again, the publication contained no clear data on adverse events.

AnticoagulantsOne RCT assessed low dose warfarin in newly inserted AV grafts, targeted to an international normalised ratio of 1.4 to 1.9 [98]. Warfarin did not decrease graft thrombosis, and the study was terminated early for major bleeding complications in the treatment group. Five patients (11 % of participants) on warfarin developed severe bleeding (including upper gastro-intestinal bleeding and cerebral haematoma). All five patients were concurrently treated with anti-platelet agents. Those in the placebo arm were had no severe bleeding.

Fish oil – omega 3 polyunsaturated fatty acidsOne systematic review compared fish oil versus placebo or no treatment, with content assessed as up-to-date to January 2017 [15]. In addition to two RCTs identified by Tanner and colleagues, this review included four additional RCTs assessing dosages ranging from 3 g three times weekly to 6 g daily. Only one study assessed the effect in AV fistulas, the others in AV grafts. In meta-analysis, the authors found moderate certainty evidence suggesting that fish oil treatment using 1.6 to 3.4 g of polyunsaturated fatty acids for 12 to 52 weeks prevented primary patency loss. There was no evidence for a differential effect between AV fistulas and grafts. It was very uncertain whether fish oil increased the risk of bleeding. No data were provided for the severity or nature of bleeding for most studies.

StatinsThere were no RCTs long-term outcomes for statins.

Vonapanitase – recombinant type I pancreatic elastase Three RCTs assessed the effect of recombinant type I pancreatic elastase applied directly to the adventitia of the AV vessels at the time of access creation [13]. At 12 months, the agent seemed to reduce the odds for AV access thrombosis, although certainty of the evidence was low due to risk of bias in the included studies and important imprecision with a confidence interval spanning the line of no effect. All three studies listed several adverse events including local symptoms secondary to the creation of a new AV access, but according to the study authors there were no significant differences between placebo and recombinant type I pancreatic elastase-treated participants. We found an additional trial that randomised 313 people to locally dripped vonapanitase or placebo [24]. At one year, the authors found very little evidence for a difference in their primary outcome – unassisted patency. Secondary patency was 13% higher in the group treated with vonapanitase, but no absolute numbers were given and risk of bias assessment hampered by the study being published as an abstract only.

Beraprost sodiumOne RCT assessed the effect of beraprost sodium, an oral synthetic analogue of prostaglandin I2, on two and one-year primary unassisted patency in 55 patients with a previously failed AV access. In 75%, the new access was an AV fistula, in the remaining quarter an AV graft. After two years, those who had received beraprost, were about three times as likely to have a patent AV access than those who had not. However, the study was considered at high risk of bias due to unclear randomising, lack of blinding, and insufficiently detailed numeric outcome reporting. The authors reported no important

of 226

bleeding events in either group, but external validity may be questioned as people considered at increased risk of bleeding had been excluded from the study.

Far infrared therapyThere was one systematic review on far infrared treatment to increase the patency of AV fistulas, with content assessed as up-to-date to January 2017 [17]. The review included 21 RCTs comprising 1899 participants at the time of access creation. Although not clearly reported, it appears all studies included people with an AV fistula. Investigators used different strategies for delivering far infrared rays; most commonly a device generating wavelengths between 5 and 25 mm, set at a height of 20 cm above the AV fistula with a treatment time of 40 min during each haemodialysis session. Most studies assessed blood volume, vessel diameter, or primary patency at various time points up to one year after fistula creation. Overall, far infrared therapy seemed to increase primary unassisted patency at 12 months, although the reliability of the evidence was compromised by the small number of studies and research groups these data were derived from, and the risk of bias in the underlying studies. In addition, five trials comprising 510 participants showed results for AV fistula occlusion rates. Overall, therapy with far infrared radiation decreased AV fistula occlusion rates. There was no evidence of heterogeneity in these trials. An older review had included only four RCTs, comprising 666 patients [16]. Primary unassisted patency was assessed in 610 patients, and far infrared therapy for 40 minutes three times weekly seemed to improve primary unassisted patency compared to controls. In addition, the two studies which reported secondary patency rates indicated a small difference in favour of far Infrared therapy. Reliability of the results was mainly limited due to high risk of bias by all being open-label, all being conducted by the same research group and all being industry sponsored.

Translation of the evidence into recommendationsThe guideline development group felt that for a positive recommendation, interventions had to improve successful use of the AV access. It was judged that in the absence of evidence for a positive effect of successful cannulation, evidence for an effect on access thrombosis would not be enough to advocate treatment. Although it is true that access thrombosis precludes successful use of the fistula for dialysis, a reduction in access thrombosis does not necessarily translate into improved patency. If these interventions, predominantly aimed at reducing platelet aggregation and coagulation, increase the risk of bleeding, a local hematoma may cause irremediable access loss. In contrast, access thrombosis may be treated with endovascular or surgical procedures whereby patency is maintained or restored. In general, there were very few studies suggesting a positive effect of a given intervention, and positive outcomes were rarely confirmed by independent sources. Often though, rather than formulating a neutral statement the group also wanted to highlight existing ambiguity by communicating the items to be weighed in decision-making.

Other guidelines on this topicESVS [33]Long-term anti-thrombotic therapy should not be used to prolong vascular access patency in haemodialysis patients. (III-C)

The Spanish Clinical Guidelines on Vascular Access for Haemodialysis [58]We suggest that antiplatelet therapy for thrombosis prophylaxis of native AV fistula be indicated on a case-by-case basis, because although it shows a decrease in the risk of thrombosis, we consider that adverse effects have not been studied with sufficient accuracy.We suggest that antithrombotic prophylaxis not be used in patients with prosthetic AV fistulas, because there is no benefit in preventing thrombosis and adverse effects have not been studied with sufficient accuracy.

UK Renal Association [92]We recommend that pharmacological and mechanical strategies are in place to maintain orrestore access patency. (1C)

of 226

CSN, KDIGO, KHA-CARI, NKF-KDOQI, or NICE provide no current recommendation on this topic.

Suggestions for future researchThe low number of studies with their short follow up, few participants, and mostly single centre approach precluded the formulation of any definitive recommendation. Many studies did not assess bleeding, infection, or obstruction of the in- or outflow. Adverse events were often not considered or inadequately reported (e.g. systemic bleeding). Any future controlled study should try and close this gap. Assessment of the evidence highlights ambiguity of “maturation” as an outcome. Some studies assess outcomes up to a year after creation of the AV access and still refer to the study as evaluating maturation. Given that principles governing maturation may be different from those maintaining long-term patency, a decision was made to assess the two outcomes separately. To allow the distinction, we required more than three months of follow up for including studies in the current chapter. To facilitate interpretation, future work should focus on establishing clear definitions for concepts that are frequently used, but currently ill-defined. Determining core outcomes and harmonizing their measurement and would facilitate comparison and interpretation of study results. Perhaps the SONG initiative, which aims to establish core outcomes in chronic kidney disease, will bring more consistency in the terminology of outcome reporting in dialysis access studies [4].

of 226

17. Chapter 9. Cannulation techniques for AV fistulas


We suggest against using area technique for cannulating AV fistulas in adults treated with haemodialysis. (2D)

We suggest using either a rope ladder or buttonhole technique for cannulating AV fistulas in adults treated with haemodialysis, and letting the choice be dependent on local expertise and AV fistula characteristics. (2D)

Advice for clinical practice Antiseptic measures and practical aspects of the cannulation procedure are important in reducing

the infection risk associated with buttonhole cannulation. AV grafts are usually only cannulated using a rope ladder technique.

Rationale

BackgroundProper cannulation of the AV access is key to its preservation and prevents vascular access related morbidity. Whereas AV grafts are only cannulated using a rope ladder technique, three different methods for puncture site selection exist for AV fistulas: area, rope ladder, and buttonhole techniques (Figure 2). Area technique refers to cannulating the same general area, but not necessarily the same point, session after session. In the rope ladder technique, the cannulator rotates needle placement sites for each dialysis along the entire length of the cannulation segment. Both these techniques require the use of sharp – cutting – needles and allow skin healing after each haemodialysis session. In the buttonhole method, haemodialysis needles are inserted at the same site, angle and depth for consecutive dialysis sessions, using blunt – non cutting – needles in a fibrotic track previously created by sharp needles. The buttonhole technique has been advocated to facilitate cannulation, decrease needling pain, reduce bleeding at the end of the haemodialysis session, and prevent aneurysm development. However, it is unclear whether these benefits really exist; how they balance with infection risk; and whether technique choice influences long term AV fistula patency.


Three systematic reviews were identified [99-101], including six reports of five RCTs comparing buttonhole with ‘control’ cannulation in AV fistulas [102-107]. The included RCTs reported on patient survival, access survival, quality of life, needling pain, infection, bleeding during or after dialysis, and aneurysm development. Outcome definition and measures varied, hampering formal meta-analysis. All studies included both incident and prevalent dialysis patients dialysed through an AV fistula. Buttonhole cannulation was compared to other – often ill-defined - cannulation techniques. One RCT defined the comparator as ‘different-site technique’, a mix of rope ladder and area cannulation not further specified [102]. Another compared buttonhole with standard or control cannulation not further specified [103]. Three RCTs compared buttonhole versus rope ladder cannulation [104-106]. In all these studies, by definition, buttonhole cannulation was performed with blunt and the comparator cannulation technique with sharp needles.

of 226

Four out of five RCTs included only in-centre haemodialysis patients [102-104, 106]. Only one included both home and in-centre patients [105]. Sample sizes were generally small, including in between 56 and 140 participants.

Patient survivalOne RCT reported eight and five deaths respectively in the groups using buttonhole and other cannulation techniques after one year (MD 7%, no statistical testing) [102]. A second RCT assessed 140 in-centre haemodialysis patients over an eight-week period reported one death in each group [106]. Its 12-month follow-up study equally reported similar number of deaths in both groups [107]. Finally, two RCTs also had similar event numbers in both groups at six months [104, 105].

Access survivalAfter one year of follow-up, one RCT had no AV fistula failures in the buttonhole group. In the group using other cannulation techniques, the median time to fistula failure was 268 days (IQR 143-292) [102]. In another RCT, the median access survival was similar for both groups (about 17 months) [107].

Primary unassisted patency and secondary patencyIn one RCT both primary unassisted and secondary patency were substantially better in the buttonhole group than in the usual practice group (73% versus 48%, no statistical testing; 100% versus 86%; p=0.005) [102]. However, a second RCT reported no difference in unassisted primary and secondary patency for buttonhole versus another technique [107].

ThrombosisOne RCT found no difference in thrombosis, the event rate was generally low [107]. A second RCT observed six cases of thrombosis with usual practice versus none with buttonhole [102]. A third RCT found one fistula thrombosis in both groups [104].

Quality of lifeOnly one RCT measured quality of life. They randomised 70 adults from multiple in-centre and home training units to either buttonhole or rope ladder cannulation. After six months, they found no difference in any of the measures between the groups [105].

Cannulation painIn a systematic review, which also included non-randomised studies, buttonhole was associated with reduced needling pain in observational studies (SMD -0.76 – 95%CI -1.38 to -20.15 SD), but not among RCTs (3 RCTs; SMD 0.34 – 95%CI -0.76 to 1.43 SD) [100]. All five included RCTs assessed patient-reported pain. None of these were blinded due to the nature of the intervention, and all studies were at high risk of detection bias. One RCT found similar pain scores for buttonhole and rope ladder cannulation over an eight-week period (median scores 1. 5 versus 1.2, p=0.57), but there was some evidence suggesting more people in the buttonhole group experienced severe pain (pain score >3) (28.6% versus 15.7%, p=0.07) [106]. Of note, the investigators only specified rope-ladder being the control group cannulation technique in the title of the paper, detailed information was missing from the methods. All participants in this study used a topical 5% lidocaine gel. It can be argued that the standard use of lidocaine decreased the pain scores, possibly explaining why a difference between cannulation techniques was not seen.In a second RCT, eight out of ten patients reported buttonhole to be less painful than their previous cannulation technique [103]. The remainder reported similar pain levels. Although randomised, the trial did not compare pain levels directly and did not provide the pain scores measured in the control group. One may also question whether the assessment time frame should not be longer than one week. A third RCT reported a median pain score of 3/10 before and 2.5/10 after six months for patients cannulated using the buttonhole technique. The rope ladder group reported pain scores of 1/10 at both time points [104]. No statistical analysis was provided. Use of local anaesthetics was allowed and could have influenced the pain scores. Patients randomised to the buttonhole technique used local anaesthetics less frequently.

of 226

A fourth RCT found people using the buttonhole technique to report similar pain scores compared with control after one year [102]. The use of local anaesthetics was similar in both groups. And finally, also a fifth RCT also found no meaningful difference in pain score between buttonhole and rope ladder at baseline or at final follow-up [105]. Fewer people in the buttonhole group used xylocaine (44% versus 76.7%, p=0.01). However, five patients using buttonhole reported site pain during dialysis (p=0.01).

InfectionsA meta-analysis, which included data from four RCTs, indicated buttonhole cannulation to be associated with a threefold increase in infectious risk compared versus other cannulation techniques (4 RCTs; RR 3.34 – 95%CI 0.91 to 12.20) [101]. Event rates varied substantially, and confidence intervals crossed the line of no effect [101]. A first RCT reported one infection - defined as erythema, redness, swelling, tenderness, exudates or pus - in 37 patients in the buttonhole group versus none in the control group during three months of follow-up [103]. The investigators did not define the cannulation technique used in the control group. A second RCT reported one infection in 28 patients in the buttonhole group during a six-month follow-up period [104]. There were no infections in the rope ladder group. In contrast, a third RCT reported two cases of bacteraemia in the control group, versus none in the buttonhole group during one year of follow-up. There were, however, two cases of local infection in the buttonhole group and none in the control group [102]. Also, a fourth study documented localized infection twice as often in those using buttonhole versus other techniques during an eight-week observation period (p<0.01) [106]. Staphylococcus aureus bacteraemia was noted once in the buttonhole group and not in the control group. An 18-month follow-up report described 12 patients using buttonhole experiencing an infection versus none with standard care (p<0.001) [107]. Three of these infections were local infections, nine Staphylococcus aureus bacteraemia. Median time to first infection was about 11 months. A fifth RCT reported four infections in 34 patients in the buttonhole group and one infection in 35 patients in the rope ladder group during a six month observation period (p= 0.11) [105]. However, the patient with the infection in the rope ladder group had been cannulated using a buttonhole technique at the time of infection.

Bleeding from cannulation sites during dialysisOne RCT reported severe bleeding (needing an astringent) at the puncture site during the haemodialysis session in 2.7% of people using buttonhole versus 4.6% using another cannulation technique. An additional 11% of people using buttonhole experienced mild bleeding, versus none in the control group [103]. The authors reported no statistical analysis and the cannulation technique in the control group was not defined. A second RCT reported 11 episodes of bleeding in two patients during dialysis in the buttonhole group, compared to 17 episodes in the control group. The number of patients affected in the control group was not reported [104].

Haemostasis after needle removalTime until haemostasis was shorter than five minutes in 54% of patients using buttonhole compared with 28% for those using the undefined control technique [103]. The four other RCTs included in the systematic review by Wong and co-workers found no appreciable differences in post-dialysis bleeding times [102, 104-106].

HaematomasOne RCT reported 19 haematomas in the buttonhole group (eight of which occurred before creation of the buttonhole tract, seven while creating the buttonhole tract and four in established tracks) compared to 27 hematomas in the rope ladder group [104]. A second RCT found fewer haematomas with buttonhole cannulation (295/1000 dialysis sessions with buttonhole versus 436/1000 dialysis sessions with standard cannulation) (p=0.03) [106]. Also, a greater proportion of patients in the standard group had at least one haematoma (36% versus 17%, p=0.01). In contrast, a third RCT reported four haematomas in 34 patients in the buttonhole group and none in the usual care group (p=0.03) [105].

Aneurysm development

of 226

One RCT assessed the average increase in maximum transverse fistula diameter based on measurements from photographs taken at baseline and after 6 months [104]. Overall, in the buttonhole group AV fistulas did not increase in size. In the rope-ladder group AV fistulas widened by 30% on average, corresponding to an absolute increase of 5 mm. A second RCT found new aneurysms in 4% of patients using a buttonhole cannulation technique and in 17% of people using the control technique [102]. Enlargements of pre-existing aneurysms were found in 23% of patients in the buttonhole group and in 67% in the control group. An aneurysm was defined as a swelling of 0.5 cm or an increase in size of 0.5 cm or more. Aneurysms were assessed based on photographs on a three-monthly basis. No studies used ultrasound to assess aneurysm formation.

Translation of the evidence into recommendations The technique used for cannulation of an AV fistula has uncertain effects on patient and access survival. RCT data are scant and contradictory, making any inference for critical outcomes quite problematic. Similarly, high certainty data for quality of life that could steer judgment in decision-making are currently not available. The supposition that the buttonhole technique causes less cannulation pain is not supported by current RCTs. However, the use of local analgesic treatment possibly influenced the extent to which pain could be objectively measured. In addition, the cannulation technique used in control groups was ill-defined for most studies. There is evidence suggesting the buttonhole technique leads to an increased risk of local and systemic infections as compared with rope ladder cannulation. However, the guideline development group felt that risk may partly be modified through appropriate antiseptic measures. There is also low certainty evidence from two studies suggesting that buttonhole cannulation causes less extensive aneurysm formation, although patency rates appear to be similar. The guideline development group felt the RCT evidence base did not allow a clear recommendation in favour of a specific cannulation technique. In the absence of such evidence, they felt their advice should incorporate a large observational study including >7000 patients, indicating area technique to be associated with poorer AV fistula survival than the other two techniques [108].The group felt it reasonable to support both rope ladder and buttonhole cannulation techniques according to centre expertise, AV fistula characteristics, and patient preference. Often, the length of the fistula cannulation segment will dictate whether to opt for buttonhole or rope ladder. The guideline development also agreed that all centres would benefit from maintaining a minimal level of experience with the different techniques within the vascular access team.From the observational data, it becomes apparent that there is large variability in how the different techniques are applied in clinical practice. A single label (buttonhole, rope ladder, area cannulation) often covers different practices, which complicates interpretation of the evidence that is available. In that perspective, the guideline development group advised to have a quality improvement program in place where outcomes of cannulation are registered and analysed at regular intervals.

Other guidelines on this topicCSN [109]For adult end-stage renal disease patients receiving intensive home haemodialysis with an AV fistula, we suggest the use of rope ladder cannulation over buttonhole cannulation unless topical antimicrobial prophylaxis is used. (2D)

For adult end-stage renal disease patients using buttonhole cannulation for intensive home haemodialysis, we suggest the use of mupirocin antibacterial cream to reduce the risk of infection. (2D)

ESVS [33]In patients with a short cannulation segment the use of the buttonhole technique should be considered over other techniques. (IIIa-C)

KHA-CARI [110]Compared to the rope ladder technique, buttonhole technique is associated with an increased risk of local and systemic infection and should not be routinely performed. (Level II evidence)

of 226

NKF-KDOQI [86] Patients with fistula access should be considered for buttonhole cannulation and for self-cannulation, the buttonhole is the preferred technique.

The Spanish Clinical Guidelines on Vascular Access for Haemodialysis [58]We recommend that the rope ladder needling technique be used as the method for cannulating a prosthetic AV fistula.We recommend that the rope ladder technique be used as the preferred method for cannulating native AV fistulas.We recommend that the buttonhole technique be reserved for cannulating tortuous or deep native AV fistulas, and/or those with an extremely short venous length.

UK Renal Association [92] We recommend that the rope-ladder and buttonhole techniques should be used for cannulationof AVF and rope-ladder for AVG. (2B)

KDIGO or NICE provide no current recommendation on this topic.

Suggestions for future researchLong term RCTs are needed comparing buttonhole with well-defined other cannulation techniques in incident haemodialysis patients. Such studies should measure pain using validated methods [111]; be adequately powered for infection, patency, and quality of life; and include detailed reporting of complications.

of 226

18. Chapter 10. Needle types for AV fistulas


We suggest using either sharp needles or plastic cannulas for cannulating AV fistulas in adults treated with haemodialysis. (2C)

We recommend using blunt needles only for buttonhole cannulation of AV fistulas in adults treated with haemodialysis. (1D)

Advice for clinical practice A quality improvement program including recording and monitoring of the needle types and

cannulation techniques alongside with AV access outcomes can help monitor quality, guide changes in cannulation practice if needed, and improve quality of vascular access care.

AV grafts are usually only cannulated using sharp steel needles.

Rationale

BackgroundSharp steel needles are routinely used for cannulating AV grafts with a rope ladder technique. In contrast, various methods are used for cannulating AV fistulas. Besides differences in location and direction of needle insertion, and differences in cannulation technique, also the shape of the needle (sharp or blunt, with or without side or back holes) and the material of the conduit (steel needle or plastic cannula) may influence AV access longevity. Most units use sharp or dull steel needles which remain in situ during the dialysis session, providing a conduit for the blood flow throughout each treatment. A sharp bevelled needle can cause trauma to the vessel if inserted or placed incorrectly. It can even perforate the AV fistula should the patient inadvertently move. In some haemodialysis units, instead, a synthetic cannula is inserted together with a sharp steel introducer which is withdrawn once the cannula sits within the vessel. The cannula is then used as the conduit during the dialysis session. Other needling systems based on synthetic materials also exist. Theoretically, these synthetic needling systems should result in fewer physical fistula injuries, but benefits and harms of these alternative systems require assessment.


Three RCTs assessing different needle designs were identified. One study, published only as an abstract, compared large-gauge hollow-bore sharp steel needles (referred to in the study as standard needles) versus Nipro SafeTouch™ sharp steel needles (referred to as safety needles) [112]. The second RCT compared sharp versus blunt steel dialysis needles using a buttonhole cannulation technique [113]. A third study compared plastic cannulas with sharp needles using rope ladder technique [114]. All three RCTs were conducted in a single centre and reported outcomes after one and six months up to one year. Sample sizes were generally small, including between 33 and 39 participants.

All included RCTs reported outcomes considered relevant to this guideline: cannulation difficulty or complications, needling pain, infection, bleeding during or after dialysis, need for interventions, and patient preference. Outcome definition and measures varied. Two studies included prevalent haemodialysis patients [112, 113], one study also included incident patients [114].

of 226

One RCT assessed a composite primary outcome of ‘acute access-associated complications’, including needle stick injuries, fistula ‘blows’ (not otherwise defined), and needle dislodgement [112]. Thirty-nine participants were randomised to standard needles or safety needles. No needle stick injuries occurred in either group. In the standard needle group, 24 infiltrations – not otherwise defined - occurred during cannulation and dialysis while 15 infiltrations occurred in the safety needle group. According to the authors, the result was not statistically significant, but no analysis was provided. A second RCT reported data after less than one month in 35 participants and 335 dialysis sessions [113]. Participants were randomised at each session to have their AV fistula cannulated using a buttonhole technique with a blunt needle or a sharp needle. Resultant important carry-over between the interventions made inference particularly challenging. Overall, among 169 AV fistula cannulations randomised to blunt needles, 12 were to be ultimately cannulated with a sharp needle due to failed cannulation. Of the 166 AV fistulas randomised to sharp needles, four were ultimately cannulated with a blunt needle because the patient refused cannulation with a sharp needle or experienced pain during cannulation. Overall, the difference in failed cannulation was not statistically significant for the downstream needle. For the upstream needle, a blunt needle resulted in cannulation failure more frequently than a sharp one (6% versus 0%; p=0.001). There were no meaningful differences in needling pain, bleeding time or infection rate between the two treatment groups [113]. A third study randomised 33 participants to having their AV fistula cannulated either with plastic cannulas or sharp needles [114]. Effects on the number of people having undergone a procedure for stenosis, thrombosis or aneurysm formation were uncertain due to wide confidence intervals and imbalance in baseline event rates. With plastic cannulas, there seemed to be 50% fewer patients experiencing complications, defined as infiltration either during cannulation or dialysis itself (RR 0.53; 95%CI 0.29 to 0.97). Again, however, certainty of the evidence was low due to sample size restrictions.

Translation of the evidence into recommendations The type of needle used for cannulation of an AV fistula has very uncertain effects on patient and access survival. RCT data are scant, making any inference for critical outcomes quite problematic. Similarly, high certainty data for quality of life that could steer judgment in decision-making are currently not available. It appears that sharp steel needles less often result in failed cannulation than blunt ones. In addition, the professed benefit of less cannulation pain with blunt steel needles in buttonhole cannulation is not supported by current RCT data. Unfortunately, those data are sparse. Only one, very small, trial tested sharp needles in AV fistulas cannulated using the buttonhole technique, and the buttonhole technique was originally described using blunt needles – the aim being not to injure the cannulation tract [113]. There is only one small RCT assessing the proposition that synthetic materials used for cannulation result in less damage to the AV fistula vessel. Again, however, sample size limitations prevent preference of one material over the other [114].

Other guidelines on this topicCSN, ESVS, GEMAV, KDIGO, NKF-KDOQI, KHA-CARI, NICE provide no current recommendation on this topic.

Suggestions for future researchAdequately powered multi-centre randomised trials assessing the benefits and harms of sharp steel needles versus other needling systems using standardised outcomes would be informative. The currently available reports do not resolve the uncertainty about long term effects and incompletely describe possible adverse events.

of 226

19. Chapter 11. Timing of intervention for AV fistula thrombosis


We suggest attempting to declot a thrombosed AV fistula in adults as soon as possible under optimal conditions and before the next haemodialysis treatment. (2D)

We suggest attempting to declot a thrombosed AV fistula in adults, even if there has been a delay of days to weeks. (2D)


Rationale

BackgroundThrombosis of the AV fistula occurs quite often (one per person every four years) and is one of the most frequent causes of access failure [115]. Thrombosis causes vessel wall inflammation and structural injury to the vascular endothelium. On the assumption that the longer a blood clot is present, the more damage it inflicts, many consider an attempt at declotting the AV fistula an emergency procedure – to be performed as quickly as possible. However, such a strategy inevitably creates logistical challenges and may inadvertently lead to worse outcomes if less experienced operators must intervene in suboptimal conditions during out-of-office hours. Understanding the trade-off is key in decision-making. In AV graft thrombosis, the blood clot is relatively inert. While it is widely agreed a timely attempt at declotting the AV graft must be made to avoid central venous catheters, thrombosis is generally not considered an emergency. Therefore, this chapter only covers AV fistula thrombosis [116].


There were no RCTs comparing the benefits and harms of earlier versus later interventions for declotting a thrombosed AV fistula.

There were four retrospective analyses assessing the effect of time to intervention on outcome of the AV fistula [117-120]. All were inherently at very high risk of bias through selection, attrition, and failing to reach the optimal information size. AV fistula outcomes were mostly reported in terms of technical success and data on primary or secondary patency were largely absent.

A first, nested case-control study, including 188 AV fistulas, indicated that surgical thrombectomy within 24 hours of diagnosing thrombosis of the AV fistula, resulted in 50% technical success [117]. If deferred to any time during the first week, only 20% of thrombectomy procedures remained successful. After that, the probability of success dropped to 10%. The investigators provided univariable comparison data only, and no attempt was made to accommodate factors influencing clinical decisions.

A second, retrospective analysis, included 59 people with thrombosis of their AV fistula who had all been referred to vascular surgery as quickly as possible [118]. Surgical thrombectomy with percutaneous transluminal angioplasty or stent placement resulted in technically successful declotting

of 226

of the AV fistula in 84% of participants treated within 6 hours of diagnosis. In those treated after that window, the procedure was successful in 74% (RR 1.14 – 95%CI 0.87 to 1.49). Again, the comparison was univariable, with data unadjusted for possible confounding variables and confidence intervals very wide.

In a third study, investigators retrospectively reviewed all 60 episodes of vascular access failure within a three-year time frame [119]. More than half were treated with thrombolysis, about a third with a combination of thrombolysis and angioplasty, and one tenth with angioplasty alone. When comparing interventions executed within 48 hours of diagnosis versus those performed after 48 hours, they found the odds of intervention failure to be about half in the group with a two-day delay versus the group in which the intervention had occurred within two days (OR 0.55 – 95%CI 0.31 to 0.99). This corresponded with an estimated 32% relative lower odds for access failure at three months (OR 0.68 – 95%CI 0.36 to 1.27). The study included both AV fistulas and grafts. Although the analysis attempted to adjust for possible confounders, interaction tests or subgroup data were not available.

Finally, a group of investigators retrospectively reported their experience with streptokinase intravascularly infused at the site of the thrombosis [120]. For the 19 participants treated within 4 days of diagnosis, the procedure was successful in 16 (84%). For the eight treated afterwards, thrombolysis was successful in three (38%) (RR 2.25 – 95%CI 0.90 to 5.61). The sample size was small, cut-offs arbitrarily chosen and analyses univariable.

Translation of the evidence into recommendations AV access failure is a common and serious complication, leading to increased temporary catheter use, access creation at multiple sites, and after many years and multiple access failures to a catastrophic inability to provide haemodialysis in some cases. Thrombosis is one of the most frequent causes of access failure and successful declotting can save the access from permanent failure. Intuitively, one would think that the earlier the intervention (surgical or radiological) is undertaken, the more likely it will result in successful access salvage, as delay can only result in clot organisation, retraction and fibrosis. Indeed, for this reason, many have considered AV access thrombosis an emergency, necessitating immediate intervention. However, the evidence to support this assumption is very sparse. There have been no randomised trials assessing the effect of increasing time-to-intervention within a reasonable time-frame on access outcome and the observational data are limited and at high risk of being biased. In addition, there may be biologic reasons for challenging the existing paradigm. Given that acute thrombosis is associated with vessel wall inflammation and endothelial injury, and such early active inflammation may be pro-thrombotic in itself, it is biologically plausible that some delay in intervention may in fact avoid rapid recurrence of thrombosis after intervention. Also, a recommendation favouring the shortest possible window for intervention may have important implications for service delivery and healthcare resources. One of the included studies assessed the causes for delay in intervention – the majority were due to a lack of interventional radiology unit availability [119]. A statement favouring rapid intervention could also inadvertently lead to worse outcomes if less experienced operators must intervene in suboptimal conditions during out-of-office hours. Finally, most cases of access thrombosis are associated with an outflow stenosis which may not be amenable to surgical treatment. Adequate imaging of the inflow and outflow should be performed and thrombectomy and stenosis treated simultaneously [121-124].In the absence of a clear understanding of the trade-offs at present, it seems reasonable that timing of the intervention weighs up different factors, including the urgency for a functioning dialysis access and the availability of optimal logistical conditions to perform the best possible intervention.

Whereas there seems to be little data to support an aim for the maximum time-to-intervention, the existing data support intervention, irrespective of the time-delay. Even after two days, 70% of procedures are still technically successful (corresponding to a three-month primary patency in 63%), and up to one week, still about one in five can technically be salvaged [117, 119]. It challenges the widely-held view that late intervention is likely to be futile. Modern mechanical thrombectomy devices could be even more effective in restoring patency several days after the thrombotic event [125,

of 226

126].

Other guidelines on this topicESVS [33] We recommend that for vascular access salvage after early thrombosis, thrombectomy and revision should be performed as soon as possible. (IC)

NKF-KDOQI [86]Thrombectomy of a fistula should be attempted as early as possible after thrombosis is detected but can be successful even after several days. (B)

The Spanish Clinical Guidelines on Vascular Access for Haemodialysis [58]We recommend priority be placed on attempting to restore the patency of potentially recoverable thrombosed AV fistulas, preferably within the first 48 h. In all cases, the priority should be to salvage the AV fistula and avoid central venous catheter placement.

UK Renal Association [92]We recommend that each centre should have facilities for surgical and radiological intervention for prompt and timely treatment of AVF/Graft stenosis; a local standard policy should be developed. (1B)

CSN, ESVS, KDIGO, KHA-CARI, or NICE provide no current recommendation on this topic.

Suggestions for future researchAn adequately powered prospective observational trial, aimed at assessing standardised outcome measures according to increasing time-to-intervention analysed as a continuous variable, could be informative for answering this question. An integrated health-economic evaluation would be required to assess the desirability of implementing the service change required to meet a set maximum time-to-intervention threshold.

of 226

20. Chapter 12. Surgical and endovascular interventions for AV access thrombosis


We suggest the choice between surgical and endovascular interventions for AV access thrombosis be defined by the condition of the patient and their vascular access, as well as local expertise, as there is no evidence one approach improves outcomes more than any other. (2B)


Rationale

BackgroundTraditionally, an occluded AV access was always treated surgically. However, over the past 20 years, endovascular techniques have increasingly been developed and are used as an alternative to salvage thrombosed vascular accesses. Most centres tend to prefer either surgery or endovascular intervention, dependent on and at the same determined by, local availability and experience with either technique. We aimed to determine which interventions - surgical or endovascular - have the best risk-benefit balance in the setting of AV access salvage, both for AV fistulas and AV grafts.


Three RCTs were identified, comparing either surgical versus endovascular intervention or different types of endovascular intervention with one another. Populations, interventions, comparators and outcomes varied across studies. All trials included participants with AV grafts. The first RCT compared the efficacy of a mechanical thrombectomy device (Amplatz™) versus conventional surgical thrombo-embolectomy for declotting 174 thrombosed AV grafts [127]. Between the 109 patients randomised to mechanical thrombectomy and the 65 individuals treated with a surgical approach, there was no difference in immediate thrombectomy success and short or long-term graft patency with successful dialysis. No extensive details of differences in minor and major adverse events between the two groups were provided. Information on sequence generation (computer-based) was given but the study remained at unclear risk for most of the other sources of bias. Of note, there was no information on whether the surgical approach included construction of a new anastomosis, i.e. proximalisation of the AV access. The second RCT was a multicentre study including 120 adults with recently thrombosed AV grafts, who were randomised to hydrodynamic thrombectomy (n=62) or pulse spray thrombolysis (n=58), both endovascular procedures [128]. No statistically significant differences were noticed in either technical success (as defined by ≥80% thrombus removal) or clinical success (technical success plus being able to provide successful dialysis) at 30 and 90 days. Similarly, there were no meaningful differences in procedure-related blood loss and early or late complications. Conversely, thrombus treatment times were shorter for thrombectomy than for thrombolysis (16.8 versus 23.4 minutes; p<0.01), suggesting that hydrodynamic thrombectomy could be useful for reducing treatment procedure time with no impact on efficacy over thrombolysis. The study was unclear with respect to selection bias as no details were provided on randomisation and allocation concealment. However, there was high risk of funding bias due to a declaration of interest of one of the co-authors. The third RCT randomised 40 patients with clotted AV grafts to a lyse-and-wait technique with 4 mg

of 226

of the tissue plasminogen activator alteplase (n=20) or to a percutaneous thrombectomy device (Arrow-Trerotola™), also both endovascular interventions [129]. In addition, 20 non-consecutive patients with thrombosed synthetic AV grafts who were randomised to undergo the lyse-and-wait technique with urokinase (250,000 U) as part of an earlier clinical study served as historical controls. The immediate anatomic success rate was 95% in both the tissue plasminogen activator lyse-and-wait and percutaneous thrombectomy device groups. The mean in-room time until restored flow was significantly lower for lyse-and-wait with tissue plasminogen activator than percutaneous thrombectomy device (10 min versus 19 min; p<0.01), although there were no differences in the mean in-room procedure time. No bleeding complications occurred in the percutaneous thrombectomy device group. Conversely, seven episodes of bleeding occurred in six patients treated with tissue plasminogen activator and four of the 20 patients undergoing lyse-and-wait with urokinase had minor puncture site bleeding during the procedure. The 3-month primary patency rates were 65%, 65%, and 60% for lyse-and-wait with tissue plasminogen activator, percutaneous thrombectomy device, and lyse-and-wait with urokinase, respectively. Given the lack of information provided, the study risk of bias remained unclear for most of the items considered.

Translation of the evidence into recommendations There is little randomised evidence available addressing this issue. The three RCTs found were mostly designed to evaluate the efficacy or superiority and safety of specific (endovascular) techniques or devices rather than comparing, more generally, surgical over endovascular approaches for AV access thrombosis. In addition, no study compared any of the available procedures in AV fistulas, all participants had AV grafts. Lastly, surgical outcomes are biased if a new anastomosis, i.e. proximalisation of the AV access is included in the surgical treatment. Observational studies suggest that thrombectomies with adjuvant treatment to correct an underlying problem result in better outcomes than endovascular intervention [130]. The appropriate comparator is surgical balloon thrombectomy (without altering the anastomosis) versus endovascular intervention. Such a study has not been conducted. This heterogeneity of procedures employed, type of interventions and comparators, and outcomes analysed prevent us from drafting definitive conclusions or recommendations favouring one approach over the other.

Other guidelines on this topicCSN [90]Correct thrombosis of an AV graft with pharmaco-mechanical or mechanical thrombolysis or surgical thrombectomy. (Grade D)

ESVS [33]Surgery or endovascular methods should be considered for treatment of late thrombosis of vascular accesses depending on the centre’s expertise. (IIa-B)Treatment of vascular access thrombosis should include perioperative diagnosis and treatment of any associated stenosis. (I-C)

The Spanish Clinical Guidelines on Vascular Access for Haemodialysis [58] We recommend an imaging test be carried out after restoring AV fistula patency, which should be performed immediately after thrombectomy to detect any possible stenoses requiring treatment.We initially recommend native AV fistula with thrombosis secondary to juxta-anastomotic stenosis be treated by surgical treatment, as long as the technique does not require central venous catheter placement.We recommend the patency of native AV fistula in thromboses not associated with juxta-anastomotic stenosis be restored by surgical treatment or by endovascular therapy, using mechanical thrombectomy or aspiration devices, if necessary.We recommend it be attempted to restore the patency of thrombosed prosthetic AV fistulas by surgical or endovascular treatment.We recommend elective intervention be performed on the dysfunctional AV fistula with significant stenosis instead of restoring after thrombosis.We recommend attempting to restore the patency of thrombosed AV fistulas rather than create a new

of 226

AV fistula and place a central venous catheter, because it is associated with lower health costs, lower hospitalisation rate and lower morbimortality.

KDIGO, NKF-KDOQI, KHA-CARI, or NICE do not provide published guidance on this topic.

Suggestions for future researchGiven the lack of evidence proving the superiority of surgical over endovascular treatment for treating fistula thrombosis, adequately powered RCTs providing data on the same type of vascular access/problem (e.g. primary or recurrent thrombosis, native fistula/grafts) and any type of procedure would be highly informative. Studies should ideally target the same core set of outcomes that should be considered essential for answering this research question, such as targeted efficacy endpoints (including but not limited to anatomic and clinical success rate at established time points, procedure duration, and long-term patency with successful dialysis) and the safety profile, particularly in terms of peri- and post-procedural bleeding. Population selection and minimization of potential biases are also crucial issues given the impossibility, due to the nature of the intervention, of eliminating performance and detection bias by blinding patients, investigators and outcome assessors.

of 226

21. TablesTable 1| Hierarchy of outcomes

Hierarchy OutcomesCritically important Death

Technique/vascular access failureMajor cardiovascular eventsMajor infectionsQuality of lifeUninterrupted dialysis sessions

Highly important Hospitalisation PainPhysical limitationsBlood flow (in AV access or in dialysis machine)

Moderately important Anxiety/distressPressures (in AV access or in dialysis machine)Dialysis adequacyRecirculation

of 226

Table 2| Method of rating the certainty of the evidence for an outcome

Step 1: Starting grade according to study design

Step 2: Lower if Step 3: Higher if Step 4: determine final grade for quality of evidence

Randomised trials = High

Observational Studies = Low

Risk of Bias-1 Serious -2 Very Serious

Inconsistency-1 Serious-2 Very Serious

Indirectness-1 Serious-2 Very Serious

Imprecision-1 Serious-2 Very Serious

Publication Bias-1 Likely-2 Very likely

Large effect+1 Large+2 Very Large

Dose response+1 Evidence of a gradient

All plausible confounding+1 Would reduce a demonstrated effect+1 Would suggest a spurious effect when results show no effect

High (four plus: ⊕⊕⊕⊕)

Moderate (three plus: ⊕⊕⊕○)

Low (two plus: ⊕⊕○○) Very Low (one plus: ⊕○○○)

Adapted from Balshem H et al. J Clin Epidemiol 2011; 46: 401-406.[131]

of 226

Table 3| Grade for the overall certainty of evidence

Grade Quality Level DefinitionA High We are confident that the true effects lies close to that of the estimates

of the effect

B Moderate The true effects are likely to be close to the estimates of the effects, but there is a possibility that they are substantially different

C Low The true effects might be substantially different from the estimates of effects

D Very low The estimates are very uncertain, and often will be far from the truth.

Adapted from Guyatt GH et al. [11]

of 226

Table 4| Implications of strong and weak recommendations for stakeholders

ImplicationsGrade Patients Clinicians Policy1 - strong‘We recommend’

Most people in your situation would want the recommended course of action, only a small proportion would not

Most patients should receive the recommended course of action

The recommendation can be adopted a as policy in most situations

2 - weak‘We suggest’

Most people in your situation would want the recommended course of action, but many would not

You should recognise that different choices will be appropriate for different patientsYou must help each patient to arrive at a management decision consistent with her or his values and preferences.

Policy making will require substantial debate and involvement of many stakeholders


of 226

Table 5| Summary of case series assessing effect of radiological endovascular interventions for non-maturing AV fistulas

Technique Target lesion N studies

N patients

Clinical success (range)

1-year primary patency (%)

1-year secondary patency (%)

Balloon angioplasty

Stenosis in draining vein or juxta-anastomotic region

14 657 43 to 97%

28 to 72% 68% to 97%

Balloon angioplasty

Stenosis in arterial inflow

2 99 91 to 98%

65 to 83% 86 to 96%

Balloon assisted maturation

Non-dilating veins

4 156 55 to 89%

- -

Endovascular accessory vein obliteration

- 1 34 65% - -

Balloon angioplasty + Endovascular accessory vein obliteration

Stenosis in draining vein or juxta-anastomotic region

6 538 78 to 92%

62% 68 to 94%

Balloon angioplasty+ stent deployment

Long segment stenosis in draining vein

1 12 100% 65% 72%

of 226

Table 6| Summary of case series assessing effect of surgical interventions for non-maturing AV fistulas

Technique Target lesion

N studies

N patients

Clinical success (range)

1-year primary patency (%)

1-year secondary patency (%)

Proximal neo-anastomosis

Stenosis in juxta-anastomotic region

2 71 90% 71 to 78%

87 to 95%

Accessory vein ligation

- 3 66 82 to 89%

- 75%

Proximal neo-anastomosis + Accessory vein ligation

Stenosis in juxta-anastomotic region

2 62 87 to 94%

68% 85 to 86%

of 226

22. Figures

Figure 1| Grade System for Grading Recommendations


of 226

Figure 2| Cannulation techniques

of 226

A: rope ladder technique; B: area technique; C: buttonhole technique.

Modified from Schmidli J et al. [33]

of 226

23. Acknowledgements

We would like to express our sincerest gratitude to all the people who devoted their time and enthusiasm to help us scope out the guideline and identify priority topics. It was key to ensuring the guideline covered the clinical problems that matter to patients, their carers, and the health care professionals for whom the document was developed. A thank you goes out to our friends from KHA-CARI, for sharing their data extraction tables on selected topics to avoid duplication of effort. We would like to acknowledge all internal reviewers for taking the time to critically read the drafts of this document and to provide us with their comments: Daniel Abramowicz, Sevcan A. Bakkaloglu, Adrian Covic, Lucia Del Vecchio, Mariusz Kusztal, Elizabeth Lindley, António Norton de Matos, Joris Rotmans, Goce Spasovski, James Tattersall and Andrzej Wiecek. We strongly believe it has contributed to the quality of the guideline and has helped maximizing its practical value. A special note of gratitude goes to the colleagues from ERA-EDTA headquarters, for their assistance in disseminating the guideline to their membership, in an attempt to attract as wide an audience as possible for participation in the external review process. Finally, we gratefully acknowledge the careful assessment of the draft guideline by external reviewers. Those external reviewers who agreed to include their name here: Saeed Ahmed, Maria Dolores Arenas, Simge Bardak, Cicily Brenenstahl, Gianni Cappelli, Eleni Chelioti, Omar Dahmani, Cristiana David, Jose Luis Del Pozo, Vasilios Devetzis, Christine Dipchand, Dimitrios Divanis, Robert Ekart, Theodoros Eleftheriadis, Victor Frajewicki, Gopalakrishna Gandikota, Davide Giunzioni, Özant Helvacı, Swapnil Hiremath, Aleksandar Jankovic, Stavros Kakkos, Hideki Kawanishi, Rumeyza Kazancioglu, Krzysztof Letachowicz, Kelvin Leung, Leandro Lucca, Jennifer Macrae, Carlo Maria Massara, Walid Omais, Hoon Suk Park, Drasko Pavlovic, Aivars Petersons, Vladimir Petkov, Bui Pham Van, Vladimir Premru, Vladimir Ryasnyanskiy, Esteban Siga, Arunkumar Subbiah, Mihály Tapolyai, Dominik Uehlinger, Semitha Utham, Katarzyna Wyskida, Dorsaf Zellema. The guideline development group will consider all the valuable comments made and, where appropriate, will incorporate suggested changes in the final document.

of 226

Supplement 1| Guideline development group area of expertise

Guideline Development Group

Julien Al Shakarchi is a vascular registrar in the West Midlands Deanery, UK, and was awarded his MD in the use of Infrared thermal imaging in vascular access from the University of Birmingham in 2017. He is a reviewer for several journals including the British Journal of Surgery and is a founder and editor of the Journal of Surgical Case Reports.

Paul Berger is a consultant vascular surgeon. He obtained a PhD on “Graft infections after surgical aortic reconstructions”. After a career in vascular and endovascular surgery, during which he (co)-authored over 40 peer reviewed articles, he obtained an MBA in healthcare delivery. He currently works as a medical consultant for a large Dutch insurance company. His main focus is strategy and innovation with a special interest in e-health.

Deirdre Cassidy is an imaging scientist with industrial and academic healthcare research experience. She has expertise in magnetic imaging (MR) contrast enhanced MR techniques, medical image analysis and processing and optimising MR imaging protocols for clinical and nonclinical studies. From April 2014 until November 2015 she was a Marie Curie Research Fellow at the University of Dundee, UK/ Guerbet, ReDVA project.

Tze Yuan Chan is a consultant vascular and interventional radiologist in the Royal Liverpool University Hospital, UK. He has a subspecialty interest in vascular intervention and vascular imaging with a focus on surveillance, maintenance and challenging vascular access.

Annemieke Dhondt is a consultant nephrologist at the Ghent University Hospital, Ghent, Belgium. She heads the chronic haemodialysis unit and has special expertise in vascular access management.

Tevfik Ecder is a consultant nephrologist and professor at Istanbul Bilim University School of Medicine, Turkey, of which he is also dean. He worked as a member of the executive committee of the Turkish Society of Nephrology between 2003 and 2011. He has also been the chairman of the Cystic Kidney Diseases Working Group of the Turkish Society of Nephrology since 2002.

Pietro Finocchiaro is a consultant nephrologist at the dialysis, nephrology and transplantation unit of the Reggio Calabria hospital, Italy. Throughout his career, he areas of expertise include performing renal biopsies and placing central venous catheters for haemodialysis. His clinical interests are mainly focused on renal imaging, particularly ultrasonography of native and transplanted kidneys, and Doppler of renal arteries and the vascular access.

Maurizio Gallieni is a consultant nephrologist and the director of nephrology and dialysis at the San Carlo Borromeo Hospital, Milano, Italy. He is affiliated with the University of Milano as researcher and aggregate professor in Nephrology. In the field of vascular access, he is the coordinating editor of the Journal of Vascular Access and has been serving the Vascular Access Society since 2009 in various roles.

Jennifer Hanko is a consultant nephrologist at Belfast Health and Social Care Trust, UK, since 2011. She has a special interest in vascular access and in interventional nephrology and works closely with renal surgeons and interventional radiologists. She has been in the Vascular Access Society of Britain and Ireland since 2012 and is currently secretary of the society.

Sam Heye is an interventional radiologist at the Jessa Hospital and Ziekenhuis Oost-Limburg in Genk, Belgium, with a special interest in vascular radiology. He obtained his PhD degree in 2012 with a

of 226

doctoral thesis entitled 'Diagnostic and percutaneous interventional techniques for prevention and treatment of dysfunctional hemodialysis fistulas and grafts.

Markus Hollenbeck is head of the renal unit and head of the KfH Dialysis centre in Bottrop, Germany. He was involved in the German Vascular Access Guideline published in 2009. He is the head of the vascular access board of the German Society of Nephrology. Together with specialists of the societies of vascular surgery, radiology and angiology he developed and started a certification for interdisciplinary vascular access centers in Germany. He is a founder of the German society of vascular access, the Interdisziplinäre Arbeitsgemeinschaft für Dialysezugänge (IAD), which focuses on interests of patients, nurses, nephrologists, interventionalists and surgeons.

Jose Ibeas is a consultant nephrologist at the nephrology department and coordinator of the vascular access programme of the Hospital Universitari Parc Taulí, Sabadell, Barcelona, Spain. He has longstanding expertise in vascular access, both clinically and scientifically. He has been director of numerous national and international courses and symposia and was co-chair of the 9th congress of the Vascular Access Society in 2015. He is the secretary of the Vascular Access Working Group of the Spanish Society of Nephrology, member of the Spanish Multidisciplinary Group on Vascular Access (GEMAV) and of the Interventional Nephrology working group of the Spanish Society of Nephrology (promoting group). He is chair of the work group for the Clinical Guidelines on Vascular Access of the GEMAV, and member of the Clinical Practice Guide on Chronic Renal Disease Group of the Spanish National Health System-Guia Salud. He is also the president-elect of the Vascular Access Society.

Nicholas Inston is a renal transplant and vascular access surgeon from Birmingham, UK. He was awarded a BMedSci in 1991 for studies into endothelial derived relaxing factor and a PhD from Birmingham University in 2006 for studies into chemokines in renal transplant rejection. He continues with research and clinical trials in vascular access and has multiple publications in the field. Past president VASBI, currently council of VAS and faculty of many vascular access meetings and societies.

Tamara Jemcov is a consultant nephrologist and the chief of the Nephrology department at the Clinical Hospital Centre Zemun in Belgrade, Serbia. She has expertise in ultrasonography. In 2013, she obtained a PhD with her thesis titled “Correlation of the Physiological, Biochemical and Morphological Parameters and Outcomes of the Native Arteriovenous Fistulas for Hemodialysis”. She has been working as a vascular access coordinator for almost 10 years at the University Clinical Centre of Serbia in Belgrade and was involved in the Serbian National Register of Vascular Access that publishes an annual report on vascular access for haemodialysis. She is a member of ERBP's vascular access guideline development group, as well as a Council member of the Vascular Access Society (VAS).

Stephanie Kershaw was a dialysis access nurse specialist at the Norfolk and Norwich University Hospital, UK, where she was also a member of the UK-wide Dialysis Access Nurse Forum which she represented on this project. She is currently retired.

Aurangzaib Khawaja is a vascular surgeon in the department of renal transplantation and dialysis access surgery part of University Hospitals Birmingham in the West Midlands of the United Kingdom. He has authored, co-authored several publications and contributed to national and international presentations of original research, systematic reviews and meta-analyses and in his field with past and ongoing participation in national and international multicentre controlled trials. He has a research interest in biostatistics with both qualitative and quantitative research methodology, he currently administers research electronic data capture (REDCap) database applications, design and setup at his renal research units instance hosted at the University of Birmingham.

Mick Kumwenda MSc FRCP (London) is a consultant nephrologist renal and diabetes centre and clinical director of general internal medicine in Glan Clwyd Hospital, UK, member of the British

of 226

Renal Society Vascular Access Special Interest Group and author of the Renal Association Vascular Access Guideline 2011 (updated in 2015).

Laura Labriola is a nephrology staff member at Cliniques Universitaires Saint-Luc, Belgium, where she is in charge of the hospital hemodialysis unit and the nephrology consultations for patients hospitalized in other units. Her topics of interest are clinical outcomes in Hemodialysis, with focus on vascular access. She published several original and review articles focusing on this topic, including a meta-analysis, and she contributed and co-authored the last EBRP guidelines concerning catheter-related blood stream infections. She also worked on UMOD-related nephropathy. She is a member of the Board of NDT-Educational and a member of the ERBP working group devoted to vascular access.

Carlo Lomonte is a chief of the nephrology unit at Miulli General Hospital, Italy. He has acted as coordinator of the Vascular Access Working Group of the Italian Society of Nephrology. He is the author of over 80 scientific articles; the main clinical and research field is the dialysis vascular access. He participates on the advisory board of a first Italian Master in vascular access and is involved in an educational project with hands-on training programmes.

Marko Malovrh is a professor of internal medicine, specialist of internal medicine and nephrology and was head of Haemodialysis Centre at University Medical Centre Ljubljana, Slovenia. He was researching morphological and functional characteristics of peripheral vessels by duplex sonography as non-invasive method before arteriovenous fistula construction. The purpose of his studies was also to find the best surgical technique and additional methods for the best arteriovenous fistula function. He was one of dedicated nephrologists performing vascular access surgery. From 1996 to 2004 he was the president of Slovenian Society of Nephrology. He was the secretary of the Vascular Access Society 2005-2007; from 2007 to 2009 he was the president of Vascular Access Society. Since 2015, he is retired.

Anna Marti i Monros is a registered nephrology nurse at Hospital General Universitario, Valencia, Spain and responsible for the home haemodialysis programme. She is a member of the Spanish Vascular Access Guidelines Group, past President of European Dialysis and Transplant Nurses Association/European Renal Care Association (EDTNA/ERCA), and DOPPS European Product Manager for EDTNA/ERCA. Her main area of interest includes haemodialysis and more specifically vascular access. She has contributed to several vascular access projects both at national and international level.

Shona Matthew is the NIHR Global Health Research Unit Manager at University of Dundee, UK. From 2013-2017, she managed the ReDVA project, a joint industry-academia research programme to overcome the scientific and technical barriers to the understanding, development and adoption of technologies to combat the significant clinical problem of the failure of renal dialysis venous access. As a previous haemodialysis patient, she is familiar with many of the challenges faced by renal patients, fuelling her interest in renal research, outreach work and guideline development.

Damian McGrogan is a course tutor for the Royal College of Surgeons England, UK, and currently completing specialist general surgical training with a renal transplant interest in Northern Ireland. He was awarded his MD in 2013 focussing on methods of predicting outcomes of arteriovenous fistula. His work includes dynamic arterial assessment using near infrared spectroscopy.

Torsten Meyer is a consultant nephrologist in the department of Medicine V - Nephrology and Hypertension - at City Hospital Braunschweig, Germany. One of his main clinical interests is planning, placement and maintenance of vascular access for haemodialysis. In this role, he is in charge of a multidisciplinary team including nephrologists, vascular surgeons, interventional radiologists for haemodialysis vascular access. In addition, he is a member of the vascular access board of the German Nephrology Society and expert for the certification of vascular access centres in Germany. He contributes to the scientific programme of national conferences for vascular access. As member of ERBP's guideline development group for vascular access his main focus is on peri- and postoperative

of 226

maintenance of AV fistulas and AV grafts.

Sotirios Mikros is a consultant nephrologist currently functioning at a managerial post at the Thriassion General Hospital in Athens, Greece. Apart from his medical qualifications, he has an academic background of Statistics and Business Administration and for the past 7 years, he has worked as a qualified nephrologist in Athens, London and Edinburgh. He received Interventional Nephrology training in Rome on scholarship by the Hellenic Renal Association. His main interest is renal replacement therapy, mainly haemodialysis and vascular access.

Nils Planken is a radiologist at the Academic Medical Center in Amsterdam, The Netherlands. After medical school and his PhD-research on "Hemodialysis vascular access imaging" at Maastricht University, he enrolled into the Radiology training programme of the Academic Medical Center in Amsterdam. He specialized in non-invasive cardiovascular imaging. He has (co)-authored over 50 peer reviewed articles and is involved in national and international clinical guidelines on non-invasive cardiovascular imaging.

Steve Powell is a consultant interventional radiologist, specialising in dialysis access interventions and surveillance. From 2011-2013 he was President of the Vascular Access Society of Britain and Ireland. He has published on percutaneous thrombectomy, Doppler surveillance of AV fistulae and the radiology of kidney transplantation.

Ramon Roca-Tey is a consultant nephrologist at the nephrology department of the Hospital de Mollet, Barcelona, Spain. His research activity focuses on the field of vascular access for hemodialysis. He has participated as co-chair, co-author and co-editor of the Spanish Clinical Guidelines of Vascular Access for Hemodialysis 2017. He has also participated as an external reviewer of the Vascular Access Clinical Practice Guidelines of the European Society of Surgery 2018. At present, he is a council member of the Vascular Access Society, coordinator of the Spanish Multidisciplinary Group on Vascular Access, coordinator of the vascular access working group of the Spanish Society of Nephrology, coordinator of the vascular access working group of the Catalan Society of Nephrology, member of the Diagnostic and Interventional Nephrology working group of the Spanish Society of Nephrology and president of the Research and Innovation Committee at the Hospital de Mollet, Barcelona, Spain. He was chair of the 9th Congress of the Vascular Access Society (Barcelona, 2015).

Rose Ross is a clinical vascular scientist with a specialist interest in the management, with a PhD investigating the role of ultrasound in surveillance. She was part of the ReDVA group a joint industry-academia research programme

Jan Tordoir is an associate professor of surgery at the department of Vascular Surgery of the Maastricht University Medical Centre, The Netherlands, and director of the Non-invasive Vascular Lab. In addition to vascular surgery, his specific interest is vascular access in dialysis patients. He has published more than 185 papers including reviews and book chapters on vascular surgery and dialysis vascular access. He is founder and past president of the European Vascular Access Society. He is chief-editor of the Journal of Vascular Access. In concordance with a multidisciplinary expert group of nephrologists, surgeons and interventional radiologists, he was involved in the creation of the European guidelines for Vascular Access. Recently, he started working on a network of centres of excellence to improve vascular access care for haemodialysis patients.

Max Troxler is a consultant vascular and trauma surgeon at Leeds Teaching Hospitals Trust, UK, where he is lead clinician for vascular and trauma surgery and surgical lead for vascular access. During his vascular surgery training he undertook a fellowship in Interventional Radiology achieving qualifications in Clinical Radiology and Ultrasound Imaging. He has a particular interest in challenging vascular access and is a member of faculty for the Royal College of Surgeons and the Vascular Society of Great Britain & Ireland's Access for Dialysis Courses.

of 226

Raymond Vanholder is a consultant nephrologist and previous head of the nephrology department at Ghent University Hospital, Belgium. Until autumn 2010, he was a member of the executive board and treasurer of Kidney Disease Improving Global Outcomes. From 2009 to 2011 he was chairman of ERBP. Before that, he coordinated the European Best Practice Guidelines on hemodialysis. He was founding president of the Belgian Society of Nephrology, is past president of the European Society of Artificial Organs and of ERA-EDTA. He is currently the chairman of the European Kidney Health Alliance and also of the European Chronic Disease Alliance.

Franks Vermassen is a vascular surgeon, heading the department of Thoracic and Vascular Surgery at Ghent University Hospital. Over the past three decades, he has held a special interest in vascular access and has developed extensive expertise both in open surgical procedures and endovascular techniques. He is a keen academic, who has published on a wide variety of topics within his field and supports a team of researchers at his department

Gunilla Welander is a consultant nephrologist at the nephrology unit Centralsjukhuset Karlstad, Sweden. Her long interest in hemodialysis and vascular access has resulted in an initiative to start a regional vascular access registry, which has become a national registry since 2011. She produces annual reports of vascular access in Sweden in collaboration with a vascular surgeon. The reports serve as a national audit of current clinical practice and outcome in vascular access. Since 2012, she has been a member of the Swedish Renal Registry steering committee. Since 2016, she has been a member of the council of Vascular Access Society.

Teun Wilmink is a consultant vascular surgeon at the Heart of England NHS foundation Trust in Birmingham, UK and an honorary senior lecturer at the University of Birmingham. As a consultant vascular surgeon, he has developed a special interest in vascular access surgery and has published widely on this subject. He has been on the board of the Vascular Access society and is an associate editor of the Journal of Vascular Access. He has an interest in surgical training and has taught widely on vascular access and is a faculty member on the Royal College of Surgeons vascular access course and the European vascular access course.


Davide Bolignano is a nephrologist and clinical researcher at the Institute of Clinical Physiology of the Italian National Council of Research, based in Reggio Calabria, Italy. He has serves ERBP as methodologist since 2011, contributing to the realization and updating of various clinical practice guidelines, position papers and systematic reviews pertaining the clinical management of renal patients. In 2018 he became member of the ERBP Advisory Board.

Christiane Drechsler is a consultant nephrologist at the University of Würzburg in Germany. She has also been trained in clinical epidemiology at the Netherlands Institute of Health Sciences in Rotterdam, and the department of Clinical Epidemiology in Leiden, the Netherlands. She graduated with a Master of Science in 2007 and with a PhD in clinical epidemiology in 2010. Her research work focuses on sudden cardiac death and the clinical epidemiology of cardiac and diabetic complications in CKD. She has published a variety of scientific papers and is a regular reviewer of scientific papers in nephrology.

Jonathan Fox is consultant nephrologist in the Glasgow Renal & Transplant Unit, UK, and an Honorary Professor in Medicine in the University of Glasgow, UK. He was Secretary-Treasurer of the ERA-EDTA from 2014-2017. He is a member of the UEMS Renal Section, a Theme Editor of NDT (Evidence-based Nephrology - Systematic Reviews - Guidelines) and the current Chair of European Renal Best Practice (ERBP).

Maria Haller is a resident in nephrology at Ordensklinikum Linz Elisabethinen, Austria. She is a senior fellow of the Methods Support Team, serving since 2012. She holds a master’s degree in

of 226

Epidemiology of the Harvard Chan School of Public Health as well as in Health Care Management of the Vienna University of Economics and Business. She has contributed to and co-authored several systematic reviews, including Cochrane reviews, as well as clinical practice guidelines produced by European Renal Best Practice. In addition, she is currently pursuing a PhD in biostatistics at the Medical University of Vienna focusing on research topics in renal transplantation.

Muguet Koobasi is an information specialist for ERBP (European Renal Best Practice), the guideline development body of ERA-EDTA (European Renal Association - European Dialysis and Transplant Association).

Evi Nagler is a consultant nephrologist at Ghent University Hospital, Belgium. She has acted as methodologist within the methods support team since 2009 and became ERBP’s Vice Chair in 2017. In that role, she has contributed to and co-authored multiple clinical practice guidelines that have been produced by the guideline body. In addition, she has contributed to several Cochrane systematic reviews. As member of the methods support team, she is primarily responsible for providing methodological support to the guideline development working groups. In addition, she is involved with process management and as such engaged in optimizing the tools and techniques used in the management of the guideline development process.

Ionut Nistor is a consultant nephrologist at the nephrology department, "Gr. T. Popa" University of Medicine and Pharmacy, Iasi, Romania. He joined ERBP from August 2011 and has served ERBP as a member of the methods support team ever since. His research interests also include cardiovascular complication in CKD patients, dialysis and transplant patients. He trained in the skills of guideline-related literature searching and evidence grading from Cochrane Kidney and Transplant. His role is to assist guideline authors in preparing and updating ERBP guidelines.

Wim Van Biesen is professor of nephrology at the Ghent University Hospital, Belgium. He is author and co-author of more than 300 articles dealing with a wide variety of topics in nephrology (peritoneal dialysis, haemodialysis, and chronic kidney disease management), intensive care nephrology. He is the former chair of ERBP. He is also Subject Editor for dialysis for Nephrology, Dialysis and Transplantation and is member of the editorial board of different other journals. He is a regular reviewer of scientific papers for different journals on nephrology, intensive care and epidemiology.

Sabine van der Veer is a Research Fellow at the Centre of Health Informatics, University of Manchester, UK. She was an ERBP fellow within the Methods Support Team from 2012-2016 and has contributed to several clinical practice guidelines. In addition to supporting the overall development process, she was responsible for scoping the current Vascular Access guideline, for which she consulted an international panel of 85 kidney patients and more than 1,000 clinicians.

of 226

Supplement 2| Declaration of interest statementsJulien Al Shakarchi 1. Do you have, or have you had during the past 2 years, any formal association with a company or other interested party?No2. Do you have, or have you had during the past 2 years, any of the following types of association with a company or other interested party?No3. Do you have, or have you had during the past 2 years, any job, position, research grant, or other grant that involved a company or other interested party?No4. Other potential conflicts of interest?No5. Is there anything else that might influence your judgement, or might be perceived to do so?No6. Member (current) of any kind of committee, board, WG, etc. of another scientific association with similar aims as ERA-EDTANo

Paul Berger1. Do you have, or have you had during the past 2 years, any formal association with a company or other interested party?Other formal associationDate:2017- 2018Company or interest group: Insurance companyValue: More than EUR 10,000Payment made to: Personal accountNature of interest: Currently employed as medical advisor2. Do you have, or have you had during the past 2 years, any of the following types of association with a company or other interested party?No3. Do you have, or have you had during the past 2 years, any job, position, research grant, or other grant that involved a company or other interested party?No4. Other potential conflicts of interest?No5. Is there anything else that might influence your judgement, or might be perceived to do so?No6. Member (current) of any kind of committee, board, WG, etc. of another scientific association with similar aims as ERA-EDTANo

Deirdre Cassidy1. Do you have, or have you had during the past 2 years, any formal association with a company or other interested party?Employee (full-time or part-time) of companyDate: 2017- 2018Company or interest group: GE HealthcareValue: More than EUR 10,000Payment made to: Personal account2. Do you have, or have you had during the past 2 years, any of the following types of association with a company or other interested party?

of 226

Conference/meeting registration fees paid or reimbursedDate: 2017- 2018Company or interest group: GE HealthcareValue: EUR 1,000-10,000Payment made to: Personal accountTravel or accommodation provided or reimbursedDate: 2017- 2018Company or interest group: GE HealthcareValue: EUR 1,000-10,000Payment made to: Personal account3. Do you have, or have you had during the past 2 years, any job, position, research grant, or other grant that involved a company or other interested party?No4. Other potential conflicts of interest?No5. Is there anything else that might influence your judgement, or might be perceived to do so?No6. Member (current) of any kind of committee, board, WG, etc. of another scientific association with similar aims as ERA-EDTANo

Tze Chan1. Do you have, or have you had during the past 2 years, any formal association with a company or other interested party?Consultant for companyDate: 2017- 2018Company or interest group: Cook MedicalValue: Less than EUR 1,000Payment made to: Personal account2. Do you have, or have you had during the past 2 years, any of the following types of association with a company or other interested party?No3. Do you have, or have you had during the past 2 years, any job, position, research grant, or other grant that involved a company or other interested party?No4. Other potential conflicts of interest?No5. Is there anything else that might influence your judgement, or might be perceived to do so?No6. Member (current) of any kind of committee, board, WG, etc. of another scientific association with similar aims as ERA-EDTANo

Annemieke Dhondt1. Do you have, or have you had during the past 2 years, any formal association with a company or other interested party?No2. Do you have, or have you had during the past 2 years, any of the following types of association with a company or other interested party?No3. Do you have, or have you had during the past 2 years, any job, position, research grant, or other grant that involved a company or other interested party?Principal investigatorDate: 2016- 2017Company or interest group: Roche

of 226

Value: More than EUR 10,000Payment made to: Hospital/institutionNature of interest: clinical trialNature of restriction: Restricted: clinical trial4. Other potential conflicts of interest?No5. Is there anything else that might influence your judgement, or might be perceived to do so?No6. Member (current) of any kind of committee, board, WG, etc. of another scientific association with similar aims as ERA-EDTANo

Tevfik Ecder1. Do you have, or have you had during the past 2 years, any formal association with a company or other interested party?No2. Do you have, or have you had during the past 2 years, any of the following types of association with a company or other interested party?No3. Do you have, or have you had during the past 2 years, any job, position, research grant, or other grant that involved a company or other interested party?No4. Other potential conflicts of interest?No5. Is there anything else that might influence your judgement, or might be perceived to do so?No6. Member (current) of any kind of committee, board, WG, etc. of another scientific association with similar aims as ERA-EDTANo

Pietro Finocchiaro1. Do you have, or have you had during the past 2 years, any formal association with a company or other interested party?No2. Do you have, or have you had during the past 2 years, any of the following types of association with a company or other interested party?No3. Do you have, or have you had during the past 2 years, any job, position, research grant, or other grant that involved a company or other interested party?No4. Other potential conflicts of interest?No5. Is there anything else that might influence your judgement, or might be perceived to do so?No6. Member (current) of any kind of committee, board, WG, etc. of another scientific association with similar aims as ERA-EDTANo

Maurizio Gallieni1. Do you have, or have you had during the past 2 years, any formal association with a company or other interested party?Consultant for companyDate: 2017- 2018Company or interest group: MedtronicValue: EUR 1,000-10,000

of 226

Payment made to: Personal accountDate: 2018- 2018Company or interest group: BBraunValue: More than EUR 10,000Payment made to: Hospital/institution2. Do you have, or have you had during the past 2 years, any of the following types of association with a company or other interested party?Travel or accommodation provided or reimbursedDate: 2018- 2018Company or interest group: AmicusValue: Less than EUR 1,000Payment made to: Other, Direct payment to provider3. Do you have, or have you had during the past 2 years, any job, position, research grant, or other grant that involved a company or other interested party?No4. Other potential conflicts of interest?No5. Is there anything else that might influence your judgement, or might be perceived to do so?No6. Member (current) of any kind of committee, board, WG, etc. of another scientific association with similar aims as ERA-EDTAYes. Vascular Access Society - Council Member

Jennifer Hanko1. Do you have, or have you had during the past 2 years, any formal association with a company or other interested party?No2. Do you have, or have you had during the past 2 years, any of the following types of association with a company or other interested party?Giving expert/scientific adviceDate: 2018- 2018Company or interest group: BardValue: Less than EUR 1,000Payment made to: Other, Value not yet determined3. Do you have, or have you had during the past 2 years, any job, position, research grant, or other grant that involved a company or other interested party?No4. Other potential conflicts of interest?No5. Is there anything else that might influence your judgement, or might be perceived to do so?No6. Member (current) of any kind of committee, board, WG, etc. of another scientific association with similar aims as ERA-EDTAYes. Vascular Access Society of Britain and Ireland

Sam Heye1. Do you have, or have you had during the past 2 years, any formal association with a company or other interested party?No2. Do you have, or have you had during the past 2 years, any of the following types of association with a company or other interested party?No3. Do you have, or have you had during the past 2 years, any job, position, research grant, or other grant that involved a company or other interested party?No

of 226

4. Other potential conflicts of interest?No5. Is there anything else that might influence your judgement, or might be perceived to do so?No6. Member (current) of any kind of committee, board, WG, etc. of another scientific association with similar aims as ERA-EDTANo

Markus Hollenbeck1. Do you have, or have you had during the past 2 years, any formal association with a company or other interested party?No2. Do you have, or have you had during the past 2 years, any of the following types of association with a company or other interested party?Giving expert/scientific adviceDate: 2017- 2018Company or interest group: DiavantisValue: EUR 1,000-10,000Payment made to: Personal accountLecturing, chairing lectures or participation in symposia/panel discussionsDate: 2017- 2017Company or interest group: B.BraunValue: Less than EUR 1,000Payment made to: Personal accountConference/meeting registration fees paid or reimbursedTravel or accommodation provided or reimbursedDate: 2017- 2017Company or interest group: GoreValue: Less than EUR 1,000Payment made to: Personal account3. Do you have, or have you had during the past 2 years, any job, position, research grant, or other grant that involved a company or other interested party?No4. Other potential conflicts of interest?No5. Is there anything else that might influence your judgement, or might be perceived to do so?No6. Member (current) of any kind of committee, board, WG, etc. of another scientific association with similar aims as ERA-EDTAYes. IAD Board (Interdisziplinäre Arbeitsgemeinschaft Dialysezugang)

Jose Ibeas1. Do you have, or have you had during the past 2 years, any formal association with a company or other interested party?Own a company/involved in start-up company in the healthcare or medical science fieldsDate: 2017- 2018Company or interest group: NephrocloudValue: Less than EUR 1,000Payment made to: Hospital/institutionHold patent related to healthcare or medical scienceDate: 2017- 2018Company or interest group: NephrocloudValue: Less than EUR 1,000Payment made to: Hospital/institution2. Do you have, or have you had during the past 2 years, any of the following types of association with

of 226

a company or other interested party?Lecturing, chairing lectures or participation in symposia/panel discussionsDate: 2017- 2017Company or interest group: AmgenValue: Less than EUR 1,000Payment made to: Personal accountDate: 2017- 2017Company or interest group: BARDValue: EUR 1,000-10,000Payment made to: Personal accountDate: 2017- 2018Company or interest group: ROVIValue: Less than EUR 1,000Payment made to: Personal account3. Do you have, or have you had during the past 2 years, any job, position, research grant, or other grant that involved a company or other interested party?No4. Other potential conflicts of interest?No5. Is there anything else that might influence your judgement, or might be perceived to do so?No6. Member (current) of any kind of committee, board, WG, etc. of another scientific association with similar aims as ERA-EDTAYes. President Elect of the Vascular Access Society, International Committee of the American Society of Diagnositc and Interventional Nephrology, Secretary of the Work Group of Vascular Access of the Spanish Society of Nephrology

Nicholas Inston1. Do you have, or have you had during the past 2 years, any formal association with a company or other interested party?No2. Do you have, or have you had during the past 2 years, any of the following types of association with a company or other interested party?No3. Do you have, or have you had during the past 2 years, any job, position, research grant, or other grant that involved a company or other interested party?No4. Other potential conflicts of interest?No5. Is there anything else that might influence your judgement, or might be perceived to do so?No6. Member (current) of any kind of committee, board, WG, etc. of another scientific association with similar aims as ERA-EDTANo

Tamara Jemcov1. Do you have, or have you had during the past 2 years, any formal association with a company or other interested party?No2. Do you have, or have you had during the past 2 years, any of the following types of association with a company or other interested party?Lecturing, chairing lectures or participation in symposia/panel discussionsDate: 2018- 2018Company or interest group: Baxter AG, SwitzerlandValue: Less than EUR 1,000

of 226

Payment made to: Personal accountConference/meeting registration fees paid or reimbursedDate: 2017- 2017Company or interest group: Eupharm, doo BeogradValue: Less than EUR 1,000Payment made to: Other, Account of the Congress organizerDate: 2017- 2018Company or interest group: Medicon doo Dec, BeogradValue: Less than EUR 1,000Payment made to: Other, account of the Congress organizerDate: 2018- 2018Company or interest group: Amicus SRB doo, SrbijaValue: Less than EUR 1,000Payment made to: Other, account of the Congress organizerTravel or accommodation provided or reimbursedDate: 2017- 2017Company or interest group: Eupharm doo BeogradValue: Less than EUR 1,000Payment made to: Other, tourist agency accountDate: 2017- 2018Company or interest group: DOO Medicon Dec, BeogradValue: Less than EUR 1,000Payment made to: Other, hotel accountDate: 2018- 2018Company or interest group: Amicus Srb doo BeogradValue: Less than EUR 1,000Payment made to: Other, travel agency account3. Do you have, or have you had during the past 2 years, any job, position, research grant, or other grant that involved a company or other interested party?No4. Other potential conflicts of interest?No5. Is there anything else that might influence your judgement, or might be perceived to do so?No6. Member (current) of any kind of committee, board, WG, etc. of another scientific association with similar aims as ERA-EDTAYes. Council member of Vascular Access Society

Stephanie Kershaw1. Do you have, or have you had during the past 2 years, any formal association with a company or other interested party?No2. Do you have, or have you had during the past 2 years, any of the following types of association with a company or other interested party?No3. Do you have, or have you had during the past 2 years, any job, position, research grant, or other grant that involved a company or other interested party?No4. Other potential conflicts of interest?No5. Is there anything else that might influence your judgement, or might be perceived to do so?No6. Member (current) of any kind of committee, board, WG, etc. of another scientific association with similar aims as ERA-EDTANo

of 226

Aurangzaib Khawaja1. Do you have, or have you had during the past 2 years, any formal association with a company or other interested party?Consultant for companyDate: 2017- 2018Company or interest group: TVA Medical, SynderMedValue: EUR 1,000-10,000Payment made to: Personal account2. Do you have, or have you had during the past 2 years, any of the following types of association with a company or other interested party?Conference/meeting registration fees paid or reimbursedDate: 2018- 2018Company or interest group: TVA MedicalValue: Less than EUR 1,000Payment made to: Hospital/institution3. Do you have, or have you had during the past 2 years, any job, position, research grant, or other grant that involved a company or other interested party?No4. Other potential conflicts of interest?No5. Is there anything else that might influence your judgement, or might be perceived to do so?No6. Member (current) of any kind of committee, board, WG, etc. of another scientific association with similar aims as ERA-EDTAYes. Vascular Access Society of Britain and Ireland

Mick Kumwenda1. Do you have, or have you had during the past 2 years, any formal association with a company or other interested party?No2. Do you have, or have you had during the past 2 years, any of the following types of association with a company or other interested party?Lecturing, chairing lectures or participation in symposia/panel discussionsDate: 2017- 2018Company or interest group: SynerMed PP LtdValue: EUR 1,000-10,000Payment made to: Other, Not paidConference/meeting registration fees paid or reimbursedDate: 2017- 2018Company or interest group: SynerMed PP LtdValue: Less than EUR 1,000Payment made to: Other, Not paidTravel or accommodation provided or reimbursedDate: 2017- 2018Company or interest group: SynerMed PP LtdValue: Less than EUR 1,000Payment made to: Other, Company sent etickets and arranged accommodation directly with hotel3. Do you have, or have you had during the past 2 years, any job, position, research grant, or other grant that involved a company or other interested party?No4. Other potential conflicts of interest?No

of 226

5. Is there anything else that might influence your judgement, or might be perceived to do so?No6. Member (current) of any kind of committee, board, WG, etc. of another scientific association with similar aims as ERA-EDTAYes. UK Renal Association

Laura Labriola1. Do you have, or have you had during the past 2 years, any formal association with a company or other interested party?No2. Do you have, or have you had during the past 2 years, any of the following types of association with a company or other interested party?Lecturing, chairing lectures or participation in symposia/panel discussionsDate: 2017- 2018Company or interest group: FreseniusValue: Less than EUR 1,000Payment made to: Hospital/institutionDate: 2018- 2018Company or interest group: AmgenValue: Less than EUR 1,000Payment made to: Hospital/institutionTravel or accommodation provided or reimbursedDate: 2017- 2017Company or interest group: BellcoValue: Less than EUR 1,000Payment made to: Hospital/institution3. Do you have, or have you had during the past 2 years, any job, position, research grant, or other grant that involved a company or other interested party?Principal investigatorDate: 2017- 2017Company or interest group: AMGENValue: EUR 1,000-10,000Payment made to: Hospital/institutionNature of interest: Randomized controlled trial MBDNature of restriction: Restricted: MBDDate: 2017- 2018Company or interest group: AstellasValue: EUR 1,000-10,000Payment made to: Hospital/institutionNature of interest: Randomized controlled trial anemia in HDNature of restriction: Restricted: HD4. Other potential conflicts of interest?No5. Is there anything else that might influence your judgement, or might be perceived to do so?No6. Member (current) of any kind of committee, board, WG, etc. of another scientific association with similar aims as ERA-EDTANo

Carlo Lomonte1. Do you have, or have you had during the past 2 years, any formal association with a company or other interested party?No2. Do you have, or have you had during the past 2 years, any of the following types of association with a company or other interested party?

of 226

No3. Do you have, or have you had during the past 2 years, any job, position, research grant, or other grant that involved a company or other interested party?No4. Other potential conflicts of interest?No5. Is there anything else that might influence your judgement, or might be perceived to do so?No6. Member (current) of any kind of committee, board, WG, etc. of another scientific association with similar aims as ERA-EDTAYes. vascular access society council member

Marko Malovrh1. Do you have, or have you had during the past 2 years, any formal association with a company or other interested party?No2. Do you have, or have you had during the past 2 years, any of the following types of association with a company or other interested party?No3. Do you have, or have you had during the past 2 years, any job, position, research grant, or other grant that involved a company or other interested party?No4. Other potential conflicts of interest?No5. Is there anything else that might influence your judgement, or might be perceived to do so?No6. Member (current) of any kind of committee, board, WG, etc. of another scientific association with similar aims as ERA-EDTA: No

Anna Marti i Monros – date 08-05-20171. Do you have, or have you had during the past 2 years, any formal association with a company or other interested party?Board of companyDate: 2016- 2016Company or interest group: Arbor Research DOPPS Steering CommitteeValue: Less than EUR 1,000Payment made to: Personal accountConsultant for companyDate: 2016- 2016Company or interest group: BAXTERValue: Less than EUR 1,000Payment made to: Personal account2. Do you have, or have you had during the past 2 years, any of the following types of association with a company or other interested party?Lecturing, chairing lectures or participation in symposia/panel discussionsDate: 2016- 2016Company or interest group: NIPROValue: Less than EUR 1,000Payment made to: Personal accountDate: 2016- 2016Company or interest group: Arbor ResearchValue: EUR 1,000-10,000Payment made to: Personal accountTravel or accommodation provided or reimbursedDate: 2016- 2016

of 226

Company or interest group: NIPROValue: Less than EUR 1,000Payment made to: Other, Conference Department3. Do you have, or have you had during the past 2 years, any job, position, research grant, or other grant that involved a company or other interested party?Other type of grantDate: 2016- 2016Company or interest group: EDTNA/ERCA DOPPS European Product ManagerValue: EUR 1,000-10,000Payment made to: Personal accountNature of interest: Coordinating data collection in DOPPS European countriesNature of restriction: Unrestricted4. Other potential conflicts of interest?No5. Is there anything else that might influence your judgement, or might be perceived to do so?No6. Member (current) of any kind of committee, board, WG, etc. of another scientific association with similar aims as ERA-EDTA: Yes. EDTNA/ERCAShona Matthew – date 28/08/20171. Do you have, or have you had during the past 2 years, any formal association with a company or other interested party?Other formal associationDate: 2016- 2017Company or interest group: Guerbet, ParisValue: More than EUR 10,000Payment made to: Other, Guerbet partially funded my researchNature of interest: I have been partially funded by Guerbet since 2012. I perform image analysis on patient images obtained using their product.2. Do you have, or have you had during the past 2 years, any of the following types of association with a company or other interested party?Other type of involvementDate: 2016- 2017Company or interest group: Guerbet, ParisValue: Less than EUR 1,000Payment made to: Other, No payment madeNature of interest: Guerbet is an industrial partner in the ReDVA project, which I project manage.Date: 2016- 2017Company or interest group: Vascular Flow Technologies, Dundee, ScotlandValue: Less than EUR 1,000Payment made to: Other, No payment madeNature of interest: Vascular Flow Technologies are an industrial partner in the ReDVA project, which I project manage3. Do you have, or have you had during the past 2 years, any job, position, research grant, or other grant that involved a company or other interested party?Principal investigatorDate: 2016- 2017Company or interest group: Guerbet, ParisValue: Less than EUR 1,000Payment made to: Other, No payment made from companyNature of interest: I am a PI in the ReDVA project, an EU FP 7 Joint industry/academia projectNature of restriction: UnrestrictedDate: 2016- 2017Company or interest group: Vascular Flow TechnologiesValue: Less than EUR 1,000Payment made to: Other, No payment made from company

of 226

Nature of interest: I am a PI in the ReDVA project, an EU FP 7 Joint industry/academia projectNature of restriction: Unrestricted4. Other potential conflicts of interest?No5. Is there anything else that might influence your judgement, or might be perceived to do so?No6. Member (current) of any kind of committee, board, WG, etc. of another scientific association with similar aims as ERA-EDTAYes. I am on the steering committee and a PI and project manager of ReDVA, an EU funded joint industry academia project looking at the challenges associated with vascular access failure

Damian McGrogan1. Do you have, or have you had during the past 2 years, any formal association with a company or other interested party?No2. Do you have, or have you had during the past 2 years, any of the following types of association with a company or other interested party?No3. Do you have, or have you had during the past 2 years, any job, position, research grant, or other grant that involved a company or other interested party?No4. Other potential conflicts of interest?No5. Is there anything else that might influence your judgement, or might be perceived to do so?No6. Member (current) of any kind of committee, board, WG, etc. of another scientific association with similar aims as ERA-EDTANo

Torsten Meyer1. Do you have, or have you had during the past 2 years, any formal association with a company or other interested party?No2. Do you have, or have you had during the past 2 years, any of the following types of association with a company or other interested party?No3. Do you have, or have you had during the past 2 years, any job, position, research grant, or other grant that involved a company or other interested party?No4. Other potential conflicts of interest?No5. Is there anything else that might influence your judgement, or might be perceived to do so?No6. Member (current) of any kind of committee, board, WG, etc. of another scientific association with similar aims as ERA-EDTANo

Sotirios Mikros1. Do you have, or have you had during the past 2 years, any formal association with a company or other interested party?No2. Do you have, or have you had during the past 2 years, any of the following types of association with a company or other interested party?No3. Do you have, or have you had during the past 2 years, any job, position, research grant, or other

of 226

grant that involved a company or other interested party?No4. Other potential conflicts of interest?No5. Is there anything else that might influence your judgement, or might be perceived to do so?No6. Member (current) of any kind of committee, board, WG, etc. of another scientific association with similar aims as ERA-EDTANo

Nils Planken1. Do you have, or have you had during the past 2 years, any formal association with a company or other interested party?No2. Do you have, or have you had during the past 2 years, any of the following types of association with a company or other interested party?No3. Do you have, or have you had during the past 2 years, any job, position, research grant, or other grant that involved a company or other interested party?No4. Other potential conflicts of interest?No5. Is there anything else that might influence your judgement, or might be perceived to do so?No6. Member (current) of any kind of committee, board, WG, etc. of another scientific association with similar aims as ERA-EDTANo

Steve Powell1. Do you have, or have you had during the past 2 years, any formal association with a company or other interested party?Employee (full-time or part-time) of companyDate: 2017- 2018Company or interest group: Rutherford DiagnosticsValue: More than EUR 10,000Payment made to: Personal account2. Do you have, or have you had during the past 2 years, any of the following types of association with a company or other interested party?No3. Do you have, or have you had during the past 2 years, any job, position, research grant, or other grant that involved a company or other interested party?No4. Other potential conflicts of interest?No5. Is there anything else that might influence your judgement, or might be perceived to do so?No6. Member (current) of any kind of committee, board, WG, etc. of another scientific association with similar aims as ERA-EDTANo

Ramon Roca-Tey1. Do you have, or have you had during the past 2 years, any formal association with a company or other interested party?No

of 226

2. Do you have, or have you had during the past 2 years, any of the following types of association with a company or other interested party?No3. Do you have, or have you had during the past 2 years, any job, position, research grant, or other grant that involved a company or other interested party?No4. Other potential conflicts of interest?No5. Is there anything else that might influence your judgement, or might be perceived to do so?No6. Member (current) of any kind of committee, board, WG, etc. of another scientific association with similar aims as ERA-EDTAYes. Vascular Access Society (VAS), Spanish Society of Nephrology (SEN), Catalan Society of Nephrology (SCN)

Rose Ross – date 25/08/20171. Do you have, or have you had during the past 2 years, any formal association with a company or other interested party?No2. Do you have, or have you had during the past 2 years, any of the following types of association with a company or other interested party?No3. Do you have, or have you had during the past 2 years, any job, position, research grant, or other grant that involved a company or other interested party?No4. Other potential conflicts of interest?No5. Is there anything else that might influence your judgement, or might be perceived to do so?No6. Member (current) of any kind of committee, board, WG, etc. of another scientific association with similar aims as ERA-EDTANo

Jan Tordoir1. Do you have, or have you had during the past 2 years, any formal association with a company or other interested party?No2. Do you have, or have you had during the past 2 years, any of the following types of association with a company or other interested party?No3. Do you have, or have you had during the past 2 years, any job, position, research grant, or other grant that involved a company or other interested party?No4. Other potential conflicts of interest?No5. Is there anything else that might influence your judgement, or might be perceived to do so?No6. Member (current) of any kind of committee, board, WG, etc. of another scientific association with similar aims as ERA-EDTANo

Max Troxler1. Do you have, or have you had during the past 2 years, any formal association with a company or other interested party?No

of 226

2. Do you have, or have you had during the past 2 years, any of the following types of association with a company or other interested party?No3. Do you have, or have you had during the past 2 years, any job, position, research grant, or other grant that involved a company or other interested party?Other position in clinical trialDate: 2017- 2018Company or interest group: Site investigator for HUMANITY trialValue: EUR 1,000-10,000Payment made to: Hospital/institutionNature of interest: Site Investigator - funds provided to pay for research nurse time/investigations as per protocolNature of restriction: UnrestrictedOther type of grantDate: 2017- 2017Company or interest group: GoreValue: EUR 1,000-10,000Payment made to: Research fundNature of interest: Educational grant of €2000: to support travelling expenses of speakers at Northern Renal Access Meeting 2017 & 2018Nature of restriction: Unrestricted4. Other potential conflicts of interest?No5. Is there anything else that might influence your judgement, or might be perceived to do so?No6. Member (current) of any kind of committee, board, WG, etc. of another scientific association with similar aims as ERA-EDTANo

Raymond Vanholder1. Do you have, or have you had during the past 2 years, any formal association with a company or other interested party?No2. Do you have, or have you had during the past 2 years, any of the following types of association with a company or other interested party?Travel or accommodation provided or reimbursedDate: 2017- 2018Company or interest group: nikishoValue: EUR 1,000-10,000Payment made to: Personal account3. Do you have, or have you had during the past 2 years, any job, position, research grant, or other grant that involved a company or other interested party?No4. Other potential conflicts of interest?No5. Is there anything else that might influence your judgement, or might be perceived to do so?No6. Member (current) of any kind of committee, board, WG, etc. of another scientific association with similar aims as ERA-EDTANo

Frank Vermassen1. Do you have, or have you had during the past 2 years, any formal association with a company or other interested party?No

of 226

2. Do you have, or have you had during the past 2 years, any of the following types of association with a company or other interested party?No3. Do you have, or have you had during the past 2 years, any job, position, research grant, or other grant that involved a company or other interested party?No4. Other potential conflicts of interest?No5. Is there anything else that might influence your judgement, or might be perceived to do so?No6. Member (current) of any kind of committee, board, WG, etc. of another scientific association with similar aims as ERA-EDTANo

Gunilla Welander1. Do you have, or have you had during the past 2 years, any formal association with a company or other interested party?No2. Do you have, or have you had during the past 2 years, any of the following types of association with a company or other interested party?No3. Do you have, or have you had during the past 2 years, any job, position, research grant, or other grant that involved a company or other interested party?No4. Other potential conflicts of interest?No5. Is there anything else that might influence your judgement, or might be perceived to do so?No6. Member (current) of any kind of committee, board, WG, etc. of another scientific association with similar aims as ERA-EDTANo

Teun Wilmink1. Do you have, or have you had during the past 2 years, any formal association with a company or other interested party?No2. Do you have, or have you had during the past 2 years, any of the following types of association with a company or other interested party?Travel or accommodation provided or reimbursedDate: 2017- 2018Company or interest group: Proteon therapeuticsValue: Less than EUR 1,000Payment made to: Personal account3. Do you have, or have you had during the past 2 years, any job, position, research grant, or other grant that involved a company or other interested party?No4. Other potential conflicts of interest?No5. Is there anything else that might influence your judgement, or might be perceived to do so?No6. Member (current) of any kind of committee, board, WG, etc. of another scientific association with similar aims as ERA-EDTANo


of 226

Davide Bolignano1. Do you have, or have you had during the past 2 years, any formal association with a company or other interested party?No2. Do you have, or have you had during the past 2 years, any of the following types of association with a company or other interested party?No3. Do you have, or have you had during the past 2 years, any job, position, research grant, or other grant that involved a company or other interested party?No4. Other potential conflicts of interest?No5. Is there anything else that might influence your judgement, or might be perceived to do so?No6. Member (current) of any kind of committee, board, WG, etc. of another scientific association with similar aims as ERA-EDTANo

Christiane Drechsler1. Do you have, or have you had during the past 2 years, any formal association with a company or other interested party?No2. Do you have, or have you had during the past 2 years, any of the following types of association with a company or other interested party?No3. Do you have, or have you had during the past 2 years, any job, position, research grant, or other grant that involved a company or other interested party?Research grantDate: 2017- 2018Company or interest group: GenzymeValue: More than EUR 10,000Payment made to: Hospital/institutionNature of interest: M. FabryNature of restriction: Restricted: research project4. Other potential conflicts of interest?No5. Is there anything else that might influence your judgement, or might be perceived to do so?No6. Member (current) of any kind of committee, board, WG, etc. of another scientific association with similar aims as ERA-EDTANo

Jonathan Fox1. Do you have, or have you had during the past 2 years, any formal association with a company or other interested party?No2. Do you have, or have you had during the past 2 years, any of the following types of association with a company or other interested party?Date: 2018- 2018Company or interest group: OmerosValue: More than EUR 1,000-10,000Payment made to: personal accountNature of interest: Membership of independent data monitoring committee3. Do you have, or have you had during the past 2 years, any job, position, research grant, or other

of 226

grant that involved a company or other interested party?No4. Other potential conflicts of interest?No5. Is there anything else that might influence your judgement, or might be perceived to do so?No6. Member (current) of any kind of committee, board, WG, etc. of another scientific association with similar aims as ERA-EDTAYes. ISRNM

Maria Haller1. Do you have, or have you had during the past 2 years, any formal association with a company or other interested party?No2. Do you have, or have you had during the past 2 years, any of the following types of association with a company or other interested party?No3. Do you have, or have you had during the past 2 years, any job, position, research grant, or other grant that involved a company or other interested party?No4. Other potential conflicts of interest?No5. Is there anything else that might influence your judgement, or might be perceived to do so?Yes. ERBP Advisory Board Member6. Member (current) of any kind of committee, board, WG, etc. of another scientific association with similar aims as ERA-EDTANo

Muguet Koobasi1. Do you have, or have you had during the past 2 years, any formal association with a company or other interested party?No2. Do you have, or have you had during the past 2 years, any of the following types of association with a company or other interested party?No3. Do you have, or have you had during the past 2 years, any job, position, research grant, or other grant that involved a company or other interested party?No4. Other potential conflicts of interest?No5. Is there anything else that might influence your judgement, or might be perceived to do so?No6. Member (current) of any kind of committee, board, WG, etc. of another scientific association with similar aims as ERA-EDTANo

Evi Nagler1. Do you have, or have you had during the past 2 years, any formal association with a company or other interested party?No2. Do you have, or have you had during the past 2 years, any of the following types of association with a company or other interested party?No3. Do you have, or have you had during the past 2 years, any job, position, research grant, or other grant that involved a company or other interested party?

of 226

No4. Other potential conflicts of interest?No5. Is there anything else that might influence your judgement, or might be perceived to do so?No6. Member (current) of any kind of committee, board, WG, etc. of another scientific association with similar aims as ERA-EDTANo

Ionut Nistor1. Do you have, or have you had during the past 2 years, any formal association with a company or other interested party?No2. Do you have, or have you had during the past 2 years, any of the following types of association with a company or other interested party?No3. Do you have, or have you had during the past 2 years, any job, position, research grant, or other grant that involved a company or other interested party?No4. Other potential conflicts of interest?No5. Is there anything else that might influence your judgement, or might be perceived to do so?No6. Member (current) of any kind of committee, board, WG, etc. of another scientific association with similar aims as ERA-EDTAYes. ERBP Advisory Board Member, ERBP Methods Support Team, Member of the Council of the Romanian Society of Nephrology

Wim Van Biesen1. Do you have, or have you had during the past 2 years, any formal association with a company or other interested party?No2. Do you have, or have you had during the past 2 years, any of the following types of association with a company or other interested party?Involvement in marketing or product developmentDate: 2017- 2018Company or interest group: FreseniusValue: Less than EUR 1,000Payment made to: Personal accountLecturing, chairing lectures or participation in symposia/panel discussionsDate: 2017- 2018Company or interest group: Fresenius, Baxter, GambroValue: Less than EUR 1,000Payment made to: Personal account3. Do you have, or have you had during the past 2 years, any job, position, research grant, or other grant that involved a company or other interested party?No4. Other potential conflicts of interest?No5. Is there anything else that might influence your judgement, or might be perceived to do so?No6. Member (current) of any kind of committee, board, WG, etc. of another scientific association with similar aims as ERA-EDTANo

of 226

Sabine van der Veer1. 1. Do you have, or have you had during the past 2 years, any formal association with a company or other interested party?No2. Do you have, or have you had during the past 2 years, any of the following types of association with a company or other interested party?No3. Do you have, or have you had during the past 2 years, any job, position, research grant, or other grant that involved a company or other interested party?No4. Other potential conflicts of interest?No5. Is there anything else that might influence your judgement, or might be perceived to do so?No6. Member (current) of any kind of committee, board, WG, etc. of another scientific association with similar aims as ERA-EDTANo

of 226

Supplement 3| Review Questions - PICOM Format

Chapter 1. Medical treatments for promoting AV access maturation

Population Patients presenting for AV access creation adults, aged adults hemodialysis of any kind Intervention Anticoagulation, antiplatelet therapy Drugs effecting on endothelial integrity/action e.g. nitrates Infrared therapy Statins Fish oilComparator No or alternative medicationComparisons 1. any drug vs no medication 2. any drug vs any other medication 3. different doses of the same medicationOutcome Core outcome measures Critical outcomes - life and death 1. Patient survival 2. Access survival 3. Maturation Critical outcomes - patient impact 1. Quality of life 2. Uninterrupted dialysis sessions 3. Hospitalisations 4. Adverse events of treatmentMethodology Systematic reviews of RCTs Randomised controlled trials

Chapter 2. Surgical and endovascular interventions for promoting AV access maturation

Population Patients undergoing AV access creation adults, aged adults hemodialysis of any kind Intervention Balloon assisted maturation Devices to preform connection btw artery and vein Surgical techniques for anastomosis Ligation of side branches

Dilation of the drainage veinType of anaesthesia: regional or local

Comparator No or alternative interventionComparisons 1. any intervention vs none 2. any intervention vs any other interventionOutcome Core outcome measures Critical outcomes - life and death

of 226

1. Patient survival 2. Access survival 3. Maturation Critical outcomes - patient impact 1. Quality of life 2. Uninterrupted dialysis sessions 3. Hospitalisations 4. Adverse events of treatmentMethodology Systematic reviews of RCTs Randomised controlled trials

Chapter 3. Surgical and radiological endovascular interventions for non-maturing AV fistulas

Population Patients after fistula creation with non-maturing fistula adults, aged adults hemodialysis of any kind Intervention Balloon assisted maturation Ligation of side branches Dilation of the drainage veinComparator No or alternative interventionComparisons 1. any intervention vs none 2. any intervention vs any other interventionOutcome Core outcome measures Critical outcomes - life and death 1. Patient survival 2. Access survival 3. Maturation Critical outcomes - patient impact 1. Quality of life 2. Uninterrupted dialysis sessions 3. HospitalisationsMethodology Systematic reviews of RCTs Randomised controlled trials

Chapter 4. Self-administered interventions for AV fistula maturation

Population Patients after AV fistula creation adults, aged adults haemodialysis of any kind Intervention Patient behaviour such as exercises tight clothes self surveillance otherComparator No or alternative particular patient behavior/education

of 226

Comparisons 1. any education/behaviour vs none 2. any education/behaviour vs any otherOutcome Core outcome measures Critical outcomes - life and death 1. Patient survival 2. Access survival 3. Maturation Critical outcomes - patient impact 1. Quality of life 2. Uninterrupted dialysis sessions 3. HospitalisationsMethodology Systematic reviews of RCTs Randomised controlled trials Cohort studies Registry studies include grey literature

Chapter 5. Perioperative prophylactic antibiotics for preventing AV access infection

Population Patients undergoing vascular access creation: AV fistula or AV graft adults, aged adults haemodialysis of any kind Intervention Antibiotics such as but not limited to -caphazolin

-vancomycin or cephradine or dicloxacillin or methicillin or netilmicin or tobramycin or lincomycin or ticarcillin or clavulanate or cefuroxime or benzylpenicillin or penicillin or cefamandole or teicoplanin or ciprofloxacin or cefoxitin

Comparator No antibiotics Different time point of administrationComparisons 1. any antibiotic versus no antibiotics 2. different time points of administration of the same antibiotic 1 hour before surgery during surgery after surgeryOutcome Core outcome measures Critical outcomes - life and death 1. Patient survival 2. Access survival Critical outcomes - patient impact 1. Quality of life 2. Major infection 3. resistance to antibiotics 4. adverse events of antibioticsMethodology Systematic reviews of RCTs Randomised controlled trialsExtra Planned subgroup analysis

of 226

AV fistulas versus grafts

Chapter 6. Timing of first cannulation

Population Patients with a recently created vascular access (fistulas and grafts) adults, aged adults haemodialysis of any kind Intervention Ultrasound criteria such as but not limited to 1. diameter of fistula vein 2. blood flow 3. depth of the fistula 4. wall thickness of fistula Time based criteria for first puncture Clinical criteria-based decision for first punctureComparator No or alternative criteriaOutcome Core outcome measures Critical outcomes - life and death 1. Patient survival 2. Access survival Critical outcomes - patient impact 1. Quality of life 2. successful first cannulation/successful dialysis at start 3. local bleeding, local pain 4. Hospitalisations 5. Need for reinterventionMethodology Systematic reviews of RCTs Randomised controlled (cluster) trials Centre level studies/registry analyses Risk prediction models Risk factor analysesExtra Planned subgroup analysis AV fistulas versus grafts

Chapter 7. Vascular access surveillance

Population Patients with a functioning AV fistula or an AV graft with or without a confirmed stenosis (defined as hemodynamically relevant) anywhere in the access

Adults, aged adultsPre-dialysis and any type haemodialysis

Intervention Surveillance (not required for inclusion)

Technical method designed to identity early stenotic lesion such as: Doppler ultrasound

Blood flow measurement Venous pressure measurement

+ Pre-emptive stenosis correction (required for inclusion) Endovascular procedures: balloon dilation with or without stenting Surgical procedures

Comparator Deferred stenosis correctionComparisons Waiting until dysfunctional access

of 226

Another surveillance method + another or the same pre-emptive correction methodNo surveillance or deferred stenosis correction

Outcome Core outcome measuresCritical outcomes - life and death

1. Patient survival/death2. Access survival/loss

Critical outcomes - patient impact 1. Quality of life2. Uninterrupted dialysis sessions3. Hospitalisation4. Adverse events of the intervention: major infection, bleeding, pain

Highly important outcomes – Pathophysiology1. Blood flow

Methodology Systematic review of RCTsRandomised controlled trials

Extra Planned subgroup analysis AV fistula versus AV graft Surveillance technique Stenosis site: in-fistula or graft, outflow or inflow Intervention type

Time frame No time limit

Chapter 8. Medical treatments for maintaining long term AV access patency

Patients Patients with an AV fistula or graft adults, aged adults predialysis and hemodialysis of any kind Intervention Antiplatelets Anticoagulation ACE-i Statins Fish oil Other (infrared therapy)Comparator no or alternative medical treatmentComparisons 1. any drug vs no drug 2. any drug vs any other drug 3. different doses of the same drugOutcome Core outcome measures Critical outcomes - life and death 1. Patient survival 2. Access survival Critical outcomes - patient impact 1. Quality of life 2. Uninterrupted dialysis sessions 3. Hospitalisations 4. adverse events of treatment

of 226

Highly important outcomes - Pathophysiology 1. Blood flowMethodology Systematic reviews of RCTs Randomised controlled trialsExtra Planned subgroup analysis site of AV access AV fistula versus AV grafts Time frame no time limit

Chapter 9. Cannulation techniques for AV fistulas

Population Patients with an AV fistula adults, aged adults haemodialysis of any kind Intervention Buttonhole cannulationComparator Rope ladder cannulation or undetermined puncture site selectionComparisons 1. button hole vs rope ladder 2. button hole vs undetermined puncture site selection/area 3. rope ladder vs undetermined puncture site selection/areaOutcome Core outcome measures Critical outcomes - life and death 1. Patient survival 2. Access survival Critical outcomes - patient impact 1. Quality of life 2. Uninterrupted dialysis sessions

3. Adverse events of intervention: bleeding and haematoma at puncture side,

infection, pain, development of aneurysm 4. HospitalisationsMethodology Systematic reviews of RCTs Randomised controlled trialsExtra Planned subgroup analysis site of access single needle vs double needle self-cannulation at home or in the hospital or cannulation by nurse/professional Time frame no time limit

Chapter 10. Needle types

Population Patients with an AV fistula or graft adults, aged adults haemodialysis of any kind Intervention Sharp steel needleComparator Alternative needle: non-steel sharp needles

of 226

1. poly-ethylene needle, synthetic, ePTFE, polyurethane 2. teflon needleComparisons 1. sharp steel needle vs any other 2. sharp steel needle vs otherOutcome Core outcome measures Critical outcomes - life and death 1. Patient survival 2. Access survival Critical outcomes - patient impact 1. Quality of life 2. Uninterrupted dialysis sessions

3. Adverse events of intervention: bleeding and haematoma at puncture site,

infection, pain, development of aneurysm 4. Hospitalisations Highly important outcomes - Pathophysiology 1. Blood flowMethodology Systematic reviews Randomised controlled trialsExtra Planned subgroup analysis site of access AV fistula versus grafts single needle vs double needle self-cannulation at home or in the hospital or cannulation by nurse/professional Time frame no time limit

Chapter 11. Timing of intervention for AV fistula thrombosis

PopulationPatients with an acute AV fistula thrombosis (defined as fully occluding thrombosis) - only fistula. No AV grafts.

adults, aged adults hemodialysis of any kind Intervention Early intervention (=within 24 hours) any intervention, open or endovascularComparator Late or no intervention (=>24 hours)Comparisons 1. any early intervention vs none 2. early intervention vs other early 3. any early intervention vs any late 4. early intervention vs late interventionOutcome Core outcome measures Critical outcomes - life and death 1. Patient survival 2. Access survival Critical outcomes - patient impact 1. Quality of life 2. Uninterrupted dialysis sessions 3. Hospitalisations

of 226

4. adverse events of intervention: infection, bleeding, pain Highly important outcomes - Pathophysiology 1. Blood flowMethodology Systematic reviews Randomised controlled trialsExtra Planned subgroup analysis site of access AV fistula versus grafts Time frame no time limit

Chapter 12. Surgical and endovascular interventions for AV access thrombosis

Patients Patients with an acute AV fistula or AV graft thrombosis adults, aged adults hemodialysis of any kind Intervention Open thrombectomy with any additional procedure (such as stent, angioplasty (balloon

and basket))Comparator Alternative intervention - any endovascular thrombectomy with any additional procedure Comparisons 1. open technique versus any other 2. any intervention versus any otherOutcome Core outcome measures Critical outcomes - life and death 1. Patient survival 2. Access survival Critical outcomes - patient impact 1. Quality of life 2. Uninterrupted dialysis sessions 3. Hospitalisations 4. Adverse events of intervention: infection, bleeding, pain Highly important outcomes - Pathophysiology 1. Blood flowMethodology Systematic reviews Randomised controlled trialsExtra Planned subgroup analysis pediatric patients predialysis versus on haemodialysis site of access AV fistula versus grafts with or without stent Time frame no time limit

of 226

Supplement 4| Search Strategies

Chapter 1. Medical treatments for promoting AV access maturationBecause chapters 1, 2, 3 and 8 covered similar topics, we developed a single strategy to cover the 4 review questions.

CENTRAL – Date 11/02/2015 and 06/04/2018#1 dialysis:ti,ab,kw (Word variations have been searched)#2 (hemodialysis or haemodialysis):ti,ab,kw (Word variations have been searched)#3 (hemodiafiltration or haemodiafiltration or hemofiltration or haemofiltration):ti,ab,kw #4 (ESRF or ESKF or ESRD or ESKD or CKF or CKD or CRF or CRD):ti,ab,kw #5 (kidney failure or kidney disease or renal failure or renal disease):ti,ab,kw #6 (predialysis or pre-dialysis):ti,ab,kw (Word variations have been searched)#7 MeSH descriptor: [Hemofiltration] explode all trees#8 MeSH descriptor: [Kidney Failure, Chronic] explode all trees#9 MeSH descriptor: [Renal Dialysis] explode all trees#10 #1-#9#11 (fistula* or avf* or shunt or shunts or graft or grafts or avg*):ti,ab,kw #12 (vascular access or venous access or dialysis access or hemodialysis access or haemodialysis

access):ti,ab,kw #13 MeSH descriptor: [Arteriovenous Fistula] explode all trees#14 MeSH descriptor: [Arteriovenous Shunt, Surgical] explode all trees#15 MeSH descriptor: [Blood Vessel Prosthesis] explode all trees#16 #11-#15#17 #10 and #16 #18 matur*:ti,ab,kw (Word variations have been searched)#19 (patency or patent):ti,ab,kw (Word variations have been searched)#20 MeSH descriptor: [Vascular Patency] explode all trees#21 ((access or fistula* or avf* or shunt or shunts or graft or grafts or AVG*) near (fail* or

malfunct* or dysfunct* or outcome* or thrombos*)):ti,ab,kw #22 MeSH descriptor: [Graft Occlusion, Vascular] explode all trees#23 MeSH descriptor: [Arteriovenous Shunt, Surgical] explode all trees and with qualifier(s):

[Adverse effects - AE]#24 MeSH descriptor: [Thrombosis] explode all trees and with qualifier(s): [Prevention & control -

PC]#25 or/#18-#24#26 #17 and #25

MEDLINE – Date 11/02/2015 and 06/04/20181. exp Renal Dialysis/2. exp Hemofiltration/3. dialysis.tw.4. (hemodialysis or haemodialysis).tw.5. (hemofiltration or haemofiltration).tw.6. (hemodiafiltration or haemodiafiltration).tw.7. exp Kidney Failure, Chronic/8. (ESRF or ESKF or ESRD or ESKD or CKF or CKD or CRF or CRD).tw.9. (predialysis or pre-dialysis).tw.10. (kidney failure or kidney disease or renal failure or renal disease).tw.11. or/1-1012. exp Arteriovenous Fistula/13. exp Arteriovenous Shunt, Surgical/

of 226

14. exp Blood Vessel Prosthesis/15. (vascular access or venous access).tw.16. (dialysis access or hemodialysis access or haemodialysis access).tw.17. (fistula* or avf* or shunt or shunts or graft or grafts or AVG*).tw.18. or/12-1719. 11 and 1820. matur*.tw.21. (patency or patent).tw.22. ((access or fistula* or avf* or shunt or shunts or graft or grafts or AVG*) adj5 (fail* or malfunct* or dysfunct* or outcome* or thrombos*)).tw.23. exp Vascular Patency/24. exp Graft Occlusion, Vascular/25. Arteriovenous Shunt, Surgical/ae [Adverse Effects]26. Thrombosis/pc [Prevention & Control]27. or/20-2628. 19 and 2729. randomized controlled trial.pt.30. controlled clinical trial.pt.31. randomized.ab.32. placebo.ab.33. clinical trials as topic/34. randomly.ab.35. trial.ti.36. or/29-3537. animals/ not (humans/ and animals/)38. 36 not 3739. 28 and 38

of 226

Chapter 2. Surgical and endovascular interventions for promoting AV access maturationBecause chapters 1, 2, 3 and 8 covered similar topics, we developed a single strategy to cover the 4 review questions. See chapter 1.

Chapter 3. Surgical and radiological endovascular interventions for non-maturing AV fistulasBecause chapters 1, 2, 3 and 8 covered similar topics, we developed a single strategy to cover the 4 review questions. See chapter 1.

Chapter 4. Self-administered interventions for AV fistula maturationMEDLINE – Date: 18/03/2015 and 06/04/20181. exp Hemofiltration/2. exp renal dialysis/3. exp Kidney failure, chronic/4. h?emodialysis.tw.5. dialysis.tw.6. h?emofilt*.tw.7. h?emodiafiltrat*.tw.8. (ESRF or ESKF or ESRD or ESKD or CKF or CKD or CRF or CRD).tw.9. (kidney failure or kidney disease or renal failure or renal disease).tw.10. (predialysis or pre-dialysis).tw.11. OR 1-1012. exp arteriovenous fistula/13. exp arteriovenous shunt, surgical/14. exp Blood vessel prosthesis/15. (vascular access or venous access or dialysis access or h?emodialysis access or HD access).tw.16. (fistula* or shunt or shunts or graft or grafts or AVF* or AVG*).tw.17. (access adj3 blood?stream).tw.18. exp Catheterization/19. exp Catheterization, central venous/20. exp Catheters, Indwelling/21. Catheter$.tw.22. Central line$.tw.23. Vascath$.tw.24. CVC*.tw.25. OR 12-2426. 11 and 2527. exp self care/28. exp self-assessment/29. exp self-evaluation programs/30. exp self-examination/31. exp diagnostic self evaluation/32. (self? and (monitor* or car* or manag* or control* or surveill* or exam* or assess* or evaluat*

or diagnos* or cannulat*)).tw.33. exp patient participation/34. exp patient preference/35. exp self-efficacy/36. exp patient care planning/37. (Plan* adj2 (care or action* or treatment*)).tw.38. (Patient* adj3 (empower* or activat* or involve* or participat*)).tw.

of 226

39. ((Shared or joint or informed or collaborative) adj2 decision?making).tw.40. exp self care/41. exp self-assessment/42. exp self-evaluation programs/43. exp self-examination/44. exp diagnostic self evaluation/45. (self? and (monitor* or car* or manag* or control* or surveill* or exam* or assess* or evaluat*

or diagnos* or cannulat*)).tw.46. exp patient participation/47. exp patient preference/48. exp self-efficacy/49. exp patient care planning/50. (Plan* adj2 (care or action* or treatment*)).tw.51. (Patient* adj3 (empower* or activat* or involve* or participat*)).tw.52. ((Shared or joint or informed or collaborative) adj2 decision?making).tw.53. ((involv* or participat*) adj5 (choice* or decision*)).tw.54. (decision adj (aid* or support or tool*)).tw.55. (Access adj2 record*).tw.56. exp patient compliance/57. exp behavior/58. exp behavior control/59. (patient* adj5 (behavi* or life?style*)).tw.60. (intervention* adj5 (behavio?r or life?style*)).tw.61. exerci*.tw.62. (program* adj3 exerci*).tw.63. (cloth* or jewel* or accessor*).tw.64. (swim* or shower* or bath*).tw.65. (adhesive* or ((adhesive adj2 (material* or tape* or bandage*)) or belt*)).tw.66. exp patient education/67. exp health education/68. exp health knowledge, attitudes, practice/69. exp counseling/70. exp inpatients/ [education]71. exp outpatients/ [education]72. exp audiovisual aids/73. (patient* adj5 (educat* or counsel* program* or instruct* or teach* or train* or inform* or

literature or leaflet* or folder* or internet or website* or forum*)).tw.74. educat*.tw.75. (intervention* adj3 education*).tw.76. (program* adj3 education*).tw.77. OR 27-8078. randomized controlled trial.pt.79. Controlled clinical trial.pt.80. randomi?ed.ab.81. placebo.ab.82. exp Clinical Trials as Topic/83. randomly.ab.84. trial.ti.85. review.pt.86. meta analysis.pt.87. (systematic* and (review* or overview*)).tw.88. meta?analy*.tw.89. meta analy*.tw.90. {or 78-89}

of 226

91. animals/ not (humans/ and animals.mp.) [mp=title, abstract, original title, name of substance word, subject heading word, keyword heading word, protocol supplementary concept word, rare disease supplementary concept word, unique identifier]

92. 90 not 9193. 26 and 77 and 92

CENTRAL – Date: 18/03/2015 and 06/04/20181. exp Hemofiltration/2. exp renal dialysis/3. exp Kidney failure, chronic/4. h?emodialysis.tw.5. dialysis.tw.6. h?emofilt*.tw.7. h?emodiafiltrat*.tw.8. (ESRF or ESKF or ESRD or ESKD or CKF or CKD or CRF or CRD).tw.9. (kidney failure or kidney disease or renal failure or renal disease).tw.10. (predialysis or pre-dialysis).tw.11. {OR 1-10}12. exp arteriovenous fistula/13. exp arteriovenous shunt, surgical/14. exp Blood vessel prosthesis/15. (vascular access or venous access or dialysis access or h?emodialysis access or HD access).tw.16. (fistula* or shunt or shunts or graft or grafts or AVF* or AVG*).tw.17. (access adj3 blood?stream).tw.18. exp Catheterization/19. exp Catheterization, central venous/20. exp Catheters, Indwelling/21. Catheter$.tw.22. Central line$.tw.23. Vascath$.tw.24. CVC*.tw.25. {OR 12-24}26. 11 and 2527. exp self care/28. exp self-assessment/29. exp self-evaluation programs/30. exp self-examination/31. exp diagnostic self evaluation/32. (self? and (monitor* or car* or manag* or control* or surveill* or exam* or assess* or evaluat*

or diagnos* or cannulat*)).tw.33. exp patient participation/34. exp patient preference/35. exp self-efficacy/36. exp patient care planning/37. (Plan* adj2 (care or action* or treatment*)).tw.38. (Patient* adj3 (empower* or activat* or involve* or participat*)).tw.39. ((Shared or joint or informed or collaborative) adj2 decision?making).tw.40. exp self care/41. exp self-assessment/42. exp self-evaluation programs/43. exp self-examination/44. exp diagnostic self evaluation/45. (self? and (monitor* or car* or manag* or control* or surveill* or exam* or assess* or evaluat*

or diagnos* or cannulat*)).tw.46. exp patient participation/

of 226

47. exp patient preference/48. exp self-efficacy/49. exp patient care planning/50. (Plan* adj2 (care or action* or treatment*)).tw.51. (Patient* adj3 (empower* or activat* or involve* or participat*)).tw.52. ((Shared or joint or informed or collaborative) adj2 decision?making).tw.53. ((involv* or participat*) adj5 (choice* or decision*)).tw.54. (decision adj (aid* or support or tool*)).tw.55. (Access adj2 record*).tw.56. exp patient compliance/57. exp behavior/58. exp behavior control/59. (patient* adj5 (behavi* or life?style*)).tw.60. (intervention* adj5 (behavio?r or life?style*)).tw.61. exerci*.tw.62. (program* adj3 exerci*).tw.63. (cloth* or jewel* or accessor*).tw.64. (swim* or shower* or bath*).tw.65. (adhesive* or ((adhesive adj2 (material* or tape* or bandage*)) or belt*)).tw.66. exp patient education/67. exp health education/68. exp health knowledge, attitudes, practice/69. exp counseling/70. exp inpatients/ [education]71. exp outpatients/ [education]72. exp audiovisual aids/73. (patient* adj5 (educat* or counsel* program* or instruct* or teach* or train* or inform* or

literature or leaflet* or folder* or internet or website* or forum*)).tw.74. educat*.tw.75. (intervention* adj3 education*).tw.76. (program* adj3 education*).tw.77. {OR 27-80}78. 26 and 77

Chapter 5. Perioperative prophylactic antibiotics for preventing AV access infectionCENTRAL – Date 07/10/2014 and 06/04/2018#1 dialysis:ti,ab,kw #2 (hemodialysis or haemodialysis):ti,ab,kw #3 (hemodiafiltration or haemodiafiltration or hemofiltration or haemofiltration):ti,ab,kw #4 (ESRF or ESKF or ESRD or ESKD or CKF or CKD or CRF or CRD):ti,ab,kw #5 (kidney failure or kidney disease or renal failure or renal disease):ti,ab,kw #6 (predialysis or pre-dialysis):ti,ab,kw #7 MeSH descriptor: [Hemofiltration] explode all trees#8 MeSH descriptor: [Kidney Failure, Chronic] explode all trees#9 MeSH descriptor: [Renal Dialysis] explode all trees#10 #1 or #2 or #3 or #4 or #5 or #6 or #7 or #8 or #9#11 (fistula* or AVF* or shunt or shunts or graft or grafts or AVG*):ti,ab,kw#12 (vascular access or venous access or dialysis access or hemodialysis access or haemodialysis

access):ti,ab,kw#13 MeSH descriptor: [Blood Vessel Prosthesis] explode all trees#14 MeSH descriptor: [Arteriovenous Fistula] explode all trees#15 MeSH descriptor: [Arteriovenous Shunt, Surgical] explode all trees

of 226

#16 #11 or #12 or #13 or #14 or #15#17 #10 and #16 #18 MeSH descriptor: [Antibiotic Prophylaxis] explode all trees#19 MeSH descriptor: [Anti-Bacterial Agents] explode all trees#20 MeSH descriptor: [Antisepsis] explode all trees#21 ((prophylaxis or prophylactic) and (antibiotic* or anti-microbial* or antimicrobial* or anti-

bacterial* or antibacterial*)):ti,ab,kw #22 cefazolin:ti,ab,kw #23 vancomycin:ti,ab,kw#24 cephradine:ti,ab,kw#25 dicloxacillin:ti,ab,kw #26 methicillin:ti,ab,kw#27 netilmicin:ti,ab,kw #28 tobramycin:ti,ab,kw #29 lincomycin:ti,ab,kw #30 ticarcillin:ti,ab,kw #31 clavulanate:ti,ab,kw #32 cefuroxime:ti,ab,kw #33 benzylpenicillin:ti,ab,kw #34 penicillin:ti,ab,kw #35 cefamandole:ti,ab,kw #36 teicoplanin:ti,ab,kw #37 ciprofloxacin:ti,ab,kw #38 cefoxitin:ti,ab,kw #39 {or #18-#38} #40 #17 and #39

MEDLINE – Date 07/10/2014 and 06/04/20181. exp Renal Dialysis/2. exp Hemofiltration/3. dialysis.tw.4. (hemodialysis or haemodialysis).tw.5. (hemofiltration or haemofiltration).tw.6. (hemodiafiltration or haemodiafiltration).tw.7. exp Kidney Failure, Chronic/8. (ESRF or ESKF or ESRD or ESKD or CKF or CKD or CKF or CRD).tw.9. (predialysis or pre-dialysis).tw.10. (kidney failure or kidney disease or renal failure or renal disease).tw.11. or/1-1012. exp Arteriovenous Fistula/13. exp Arteriovenous Shunt, Surgical/14. exp Blood Vessel Prosthesis/15. (vascular access or venous access).tw.16. (dialysis access or hemodialysis or haemodialysis).tw.17. (fistula* or avf* or shunt or shunts or graft or grafts or avg*).tw.18. or/12-1719. 11 and 1820. exp Antibiotic Prophylaxis/21. exp Anti-Bacterial Agents/22. exp Antisepsis/23. ((prophylaxis or prophylactic) and (antibiotic* or anti-microbial* or antimicrobial* or anti-bacterial* or antibacterial*)).tw.24. cefazolin.tw.25. vancomycin.tw.26. cephradine.tw.

of 226

27. dicloxacillin.tw.28. methicillin.tw.29. netilmicin.tw.30. tobramycin.tw.31. lincomycin.tw.32. ticarcillin.tw.33. clavulanate.tw.34. cefuroxime.tw.35. benzylpenicillin.tw.36. penicillin.tw.37. cefamandole.tw.38. teicoplanin.tw.39. ciprofloxacin.tw.40. cefoxitin.tw.41. or/20-4042. 19 and 4143. randomized controlled trial.pt.44. controlled clinical trial.pt.45. randomized.ab.46. placebo.ab.47. clinical trials as topic/48. randomly.ab.49. trial.ti.50. or/43-4951. animals/ not (humans/ and animals/)52. 50 not 5153. 42 and 52

Chapter 6. Timing of first cannulationCENTRAL – Date 07/10/2014 and 06/04/2018#1 dialysis:ti,ab,kw#2 (hemodialysis or haemodialysis):ti,ab,kw #3 (hemodiafiltration or haemodiafiltration or hemofiltration or haemofiltration):ti,ab,kw #4 MeSH descriptor: [Renal Dialysis] explode all trees#5 MeSH descriptor: [Hemofiltration] explode all trees#6 #1 or #2 or #3 or #4 or #5#7 (fistula* or shunt or shunts or AVF* or graft or grafts or AVG*):ti,ab,kw #8 (vascular access or venous access or dialysis access or hemodialysis access or haemodialysis

access):ti,ab,kw#9 MeSH descriptor: [Blood Vessel Prosthesis] explode all trees#10 MeSH descriptor: [Arteriovenous Fistula] explode all trees#11 MeSH descriptor: [Arteriovenous Shunt, Surgical] explode all trees#12 {or #7-#11} #13 #6 and #12 #14 ((tim* or week* or month*) and (cannulat* or punctur* or matur* or adequa*)):ti,ab,kw#15 #13 and #14

MEDLINE – Date 07/10/2014 and 06/04/20181. exp Renal Dialysis/2. exp Hemofiltration/3. dialysis.tw.4. (hemodialysis or haemodialysis).tw.5. (hemofiltration or haemofiltration).tw.6. (hemodiafiltration or haemodiafiltration).tw.

of 226

7. or/1-68. exp Arteriovenous Fistula/9. exp Arteriovenous Shunt, Surgical/10. exp Blood Vessel Prosthesis/11. (vascular access or venous access).tw.12. (dialysis access or hemodialysis access or haemodialysis access).tw.13. (fistula* or avf* or shunt or shunts or graft or grafts or avg*).tw.14. or/8-1315. 7 and 1416. ((tim* or week* or month*) and (cannulat* or punctur* or matur* or adequa*)).tw.17. 15 and 16

Chapter 7. Vascular access surveillanceIn a step down hierarchical search strategy, a Cochrane review, which was on topic and up to date was identified from the Cochrane Library.

Chapter 8. Medical treatments for maintaining long term AV access patencySee chapter 1

Chapter 9. Cannulation techniques for AV fistulasCENTRAL – Date 07/10/2014 and 06/04/2018#1 dialysis:ti,ab,kw#2 (hemodialysis or haemodialysis):ti,ab,kw #3 (hemodiafiltration or haemodiafiltration or hemofiltration or haemofiltration):ti,ab,kw #4 MeSH descriptor: [Renal Dialysis] explode all trees#5 MeSH descriptor: [Hemofiltration] explode all trees#6 #1 or #2 or #3 or #4 or #5 or #6 #7 (fistula* or shunt or shunts or AVF* or graft or grafts or AVG*):ti,ab,kw #8 (vascular access or venous access or dialysis access or hemodialysis access or haemodialysis

access):ti,ab,kw #9 MeSH descriptor: [Blood Vessel Prosthesis] explode all trees#10 MeSH descriptor: [Arteriovenous Fistula] explode all trees#11 MeSH descriptor: [Arteriovenous Shunt, Surgical] explode all trees#12 {or #7-#11} #13 #6 and #12 #14 (rope ladder or rope-ladder):ti,ab,kw #15 (buttonhole or button hole or button-hole):ti,ab,kw #16 (constant-site or constant site):ti,ab,kw #17 cannula*:ti,ab,kw #18 needl*:ti,ab,kw #19 punctur*:ti,ab,kw #20 MeSH descriptor: [Punctures] explode all trees#21 MeSH descriptor: [Needles] explode all trees#22 ((sharp or dull or metal or steel or plastic or synthetic or ePTFE or polyurethan*) and (needl*

or cannula* or introducer)):ti,ab,kw #23 {or #14-#22} #24 #13 and #23

MEDLINE – Date 07/10/2014 and 06/04/20181. exp Renal Dialysis/2. exp Hemofiltration/

of 226

3. dialysis.tw.4. (hemodialysis or haemodialysis).tw.5. (hemofiltration or haemofiltration).tw.6. (hemodiafiltration or haemodiafiltration).tw.7. or/1-68. exp Arteriovenous Fistula/9. exp Arteriovenous Shunt, Surgical/10. exp Blood Vessel Prosthesis/11. (vascular access or venous access).tw.12. (dialysis access or hemodialysis access or haemodialysis access).tw.13. (fistula* or avf* or shunt or shunts or graft or grafts or avg*).tw.14. or/8-1315. 7 and 1416. (rope ladder or rope-ladder).tw.17. (buttonhole or button-hole or botton hole).tw.18. (constant-site or constant site).tw.19. cannula*.tw.20. needl*.tw.21. punctur*.tw.22. exp Punctures/23. exp Needles/24. ((sharp or dull or metal or steel or plastic or synthetic or ePTFE or polyurethan*) and (needl* or cannula* or introducer)).tw.25. or/16-2426. 15 and 2527. randomized controlled trial.pt.28. controlled clinical trial.pt.29. randomized.ab.30. placebo.ab.31. clinical trials as topic/32. randomly.ab.33. trial.ti.34. or/27-3335. animals/ not (humans/ and animals/)36. 34 not 3537. 26 and 36

Chapter 10. Needle typesSee chapter 9

Chapter 11. Timing of intervention for AV fistula thrombosisMEDLINE – Date 21/02/2016 and 06/04/20181. arteriovenous.tw2. AV.tw3. autologous.tw4. autogenous.tw5. native.tw6. dialysis.tw7. (hemodialysis or haemodialysis).tw8. access.tw9. OR/1-810. fistula.tw11. graft.tw12. shunt.tw

of 226

13. vascular.tw AND acces*.tw14. OR/10-1315. thrombectomy.tw16. embolectomy.tw17. thrombolysis.tw18. salvage.tw19. revascularization.tw20. OR/15-1921. thrombosis.tw22. thrombosed.tw23. occlusion.tw24. occluded.tw25. failed.tw26. OR/21-2527. time.tw28. timing.tw29. clinical success.tw30. technical success.tw31. patency.tw32. patent.tw33. OR/27-3234. 9 AND 14 AND 20 AND 26 AND 33

EMBASE – Date 21/02/2016 and 06/04/20181. arteriovenous.tw2. AV.tw3. autologous.tw4. autogenous.tw5. native.tw6. dialysis.tw7. (hemodialysis or haemodialysis).tw8. access.tw9. OR/1-810. fistula.tw11. graft.tw12. shunt.tw13. vascular.tw AND acces*.tw14. OR/10-1315. thrombectomy.tw16. embolectomy.tw17. thrombolysis.tw18. salvage.tw19. revascularization.tw20. OR/15-1921. thrombosis.tw22. thrombosed.tw23. occlusion.tw24. occluded.tw25. failed.tw26. OR/21-2527. time.tw28. timing.tw29. clinical success.tw30. technical success.tw31. patency.tw

of 226

32. patent.tw33. OR/27-3234. 9 AND 14 AND 20 AND 26 AND 33

WebofScience – Date 21/02/2016(((arteriovenous OR AV OR autologous OR autogenous OR native OR dialysis OR hemodialysis OR haemodialysis OR access OR acces) AND (fistula* OR graft* OR shunt* OR (vascular AND (access OR acces))))) AND (thrombectomy OR embolectomy OR thrombolysis OR salvage OR revascularization) AND (thrombosis OR thrombosed OR occlusion OR occluded OR failed) AND ((time OR timing) OR ((clinical OR technical) AND (success)) OR patency OR patent)

Chapter 12. Surgical and endovascular interventions for AV access thrombosisMEDLINE – Date 21/02/2016 (original date 24/11/2014) and 06/04/201835. arteriovenous.tw36. AV.tw37. autologous.tw38. autogenous.tw39. native.tw40. dialysis.tw41. (hemodialysis or haemodialysis).tw42. access.tw43. OR/1-844. fistula.tw45. graft.tw46. shunt.tw47. vascular.tw AND acces*.tw48. OR/10-1349. thrombectomy.tw50. embolectomy.tw51. thrombolysis.tw52. salvage.tw53. revascularization.tw54. OR/15-1955. thrombosis.tw56. thrombosed.tw57. occlusion.tw58. occluded.tw59. failed.tw60. OR/21-2561. time.tw62. timing.tw63. clinical success.tw64. technical success.tw65. patency.tw66. patent.tw67. OR/27-3268. 9 AND 14 AND 20 AND 26 AND 33

EMBASE – Date 21/02/2016 and 06/04/2018

35. arteriovenous.tw36. AV.tw37. autologous.tw

of 226

38. autogenous.tw39. native.tw40. dialysis.tw41. (hemodialysis or haemodialysis).tw42. access.tw43. OR/1-844. fistula.tw45. graft.tw46. shunt.tw47. vascular.tw AND acces*.tw48. OR/10-1349. thrombectomy.tw50. embolectomy.tw51. thrombolysis.tw52. salvage.tw53. revascularization.tw54. OR/15-1955. thrombosis.tw56. thrombosed.tw57. occlusion.tw58. occluded.tw59. failed.tw60. OR/21-2561. time.tw62. timing.tw63. clinical success.tw64. technical success.tw65. patency.tw66. patent.tw67. OR/27-3268. 9 AND 14 AND 20 AND 26 AND 33

WebofScience – Date 21/02/2016(((arteriovenous OR AV OR autologous OR autogenous OR native OR dialysis OR hemodialysis OR haemodialysis OR access OR acces) AND (fistula* OR graft* OR shunt* OR (vascular AND (access OR acces))))) AND (thrombectomy OR embolectomy OR thrombolysis OR salvage OR revascularization) AND (thrombosis OR thrombosed OR occlusion OR occluded OR failed) AND ((time OR timing) OR ((clinical OR technical) AND (success)) OR patency OR patent)

of 226

Supplement 5| Study selection flow diagrams

Chapter 1. Medical treatments for promoting AV access maturationBecause chapters 1, 2, 3 and 8 covered similar topics, we developed a single strategy to cover the 4 review questions. Search results were screened for separate inclusion in the various chapters. Some studies – typically systematic reviews could cover more than one question. c

of 226

444 Duplicates removed

31 Documents included in qualitative and quantitative analysis

13 Chapter 118 Chapter 20 Chapter 37 Chapter 8 (7 reports shared with

Chapter 1)

113 Citations excluded based on full-text screening

24 wrong study design14 wrong population21 wrong intervention or

comparator 23 duplicates31 RCTs included in SR

144 Potentially relevant citations identified for full-text review

1570 Citations excluded on screening of title and abstracts

1714 Potentially relevant citations identified for title and abstract review

1 Additional citation identified via other sources2157 Citations identified by

electronic database searching1140 Cochrane CENTRAL1017 MEDLINE

Chapter 2. Surgical and endovascular interventions for promoting AV access maturationBecause chapters 1, 2, 3 and 8 covered similar topics, we developed a single strategy to cover the 4 review questions. Search results were screened for separate inclusion in the various chapters. Some studies – typically systematic reviews could cover more than one question. See chapter 1

Chapter 3. Surgical and radiological endovascular interventions for non-maturing AV fistulasBecause chapters 1, 2, 3 and 8 covered similar topics, we developed a single strategy to cover the 4 review questions. Search results were screened for separate inclusion in the various chapters. Some studies – typically systematic reviews could cover more than one question. See chapter 1


of 226



4 Chapter 42 Chapter 7


4 wrong population26 wrong intervention or

comparator 9 duplicates




4 Additional citation identified via other sources

899 Citations identified by electronic database searching494 Cochrane Central405 MEDLINE


of 226


3 Reports of 2 studies included

26 Citations excluded based on full-text screening due to wrong intervention or comparator





232 Citations identified by electronic database searching

44 Cochrane Central188 MEDLINE


of 226




197 wrong intervention or comparator

37 duplicates1 not adults






Chapter 7. Vascular access surveillanceIn a step down hierarchical search strategy, a Cochrane review, which was on topic and up to date was identified from the Cochrane Library.

Chapter 8. Medical treatments for maintaining long term AV access patencySee chapter 1


of 226




5 wrong study design1 wrong intervention or







Chapter 10. Needle types

of 226












of 226

1 Systematic review by collaborating authors


1 Document included


79 no time comparison61 wrong population17 wrong intervention 45 wrong design5 no outcomes in time2 spanish1 not accessible




4870 Citations identified by electronic database searching


of 226









1408 Citations identified by electronic database searching 2994 additional citations identified

through manual searching

Supplement 6| Summary evidence tables

See separate document prepared in Microsoft Excel (2010)

Chapter 1. Medical treatments for promoting AV access maturationAuthorTitlePublication YearJournal

Study Design Patient Characteristics Inclusion

Comparator Intervention Amstar Score N Studies Included

N Patients Included

Age Outcome Outcome Defined As

TannerMedical adjuvant treatment to increase patency of arteriovenous fistulae and grafts2015Cochrane Database of Systematic Reviews[13]

Systematic review

InclusionRCTs of ESKD patients undergoing haemodialysis via an AV fistula or AV graftExclusionperitoneal dialysis patients, non RCTs

No adjuvant therapy

Adjuvant therapy

8/11 - Duplicate study selection and data extraction not entirely independent; status of publication possibly inclusion criterion, and misinterpretation of lack of evidence for evidence of no effect

15 2230 18 years-80 years

Graft/ fistula patency

Thrombosis

Time Point Of The Outcome

Comparison Outcome Value Comparator Group

Outcome Value Intervention Group

Analysed Participants Comparator Group

Analysed Participants Intervention Group

Effect Measure

Value Effect Measure

Lower Bound Confidence Interval

Upper Bound Confidence Interval

Evidence Certainty

1 to 18 months after AV access creation

Aspirin vs placebo 35 18 91 84 Odds Ratio

0.4 0.07 2.25 Very low: 3 studies from before 1995, at unclear risk of bias, serious inconsistency I²=79%; indirectness (intermediate for access loss) and very serious imprecision

1 month Ticlopidine vs placebo

38 20 168 171 Odds Ratio

0.45 0.25 0.82 Low: 3 studies from before 1998 at high or unclear risk of bias, and serious imprecision, sample size far below optimal information size

Not reported Dipyridamole vs placebo

6 4 19 23 Odds Ratio

0.46 0.11 1.94 Low: 1 study from 1994 at unclear risk of selection and selective reporting bias, very serious imprecision with confidence interval spanning both important decrease and increase in odds for thrombosis

Up to 37 months after AV graft insertion

Warfarin vs placebo


1.76 0.75 3.99 Low: 1 study, point estimate favouring placebo, but very imprecise.

12 months Fishoil vs placebo 54 35 109 111 Odds 0.24 0.03 1.95 Very low: 2 studies at high risk of bias, high degree of

of 226

Ratio inconsistency and very serious imprecision6 to 8 weeks after AV fistula formation

Clopidogrel vs placebo

94 55 477 482 Odds Ratio

0.4 0.13 1.19 Low: 2 studies at risk of bias, and serious imprecision with the confidence interval spanning the line of no effect

3 months Sulphinpyrazone vs placebo

2 1 8 8 Odds Ratio

0.43 0.03 5.98 Very low: one very small study from 1977 at high risk of bias, with very imprecise effect estimate.

12 months PRT-201 vs placebo

28 48 100 206 Odds Ratio

0.75 0.42 1.32 Very low: 3 studies, some at risk of bias and with very serious imprecision and confidence interval spanning line of no effect.

AuthorTitlePublication YearJournal

Study Design Location Patient Characteristics Inclusion

Comparator - TypeIntervention - Type

Amstar Score

A Priori Design Provided?

Duplicate Study Selection/ Extraction?

Comprehensive Literature Search?

Publication Inclusion Criterion?

PalmerAntiplatelet agents for chronic kidney disease2013Database of Systematic Reviews[14]

Systematic review

Global RCTs of CKD patients, including those who needed renal replacement therapy (dialysis)

Comparatorplacebo or no treatmentInterventionantiplatelet therapy

10/11 Yes Yes Yes Yes

List of in- & excluded studies?

Characteristics included studies given?

Scientific quality studies assessed?

Scientific quality studies used?

Methods to combine findings correct?

Likelihood publication bias assessed?

Conflict of interest stated?

N citations evaluated for inclusion?

N full papers evaluated for inclusion?

N studies enrolled?

N Patients Enrolled?

Yes Yes Yes Yes Yes Yes, but not for vascular outcome

No 1093 182 50 27139

of 226


Study Design Patient Characteristics Inclusion Comparator - TypeIntervention - Type

Amstar Score





List Of In- & Excluded Studies?

PalmerAntiplatelet therapy to prevent hemodialysis vascular access failure: systematic review and meta-analysis2013American Journal of Kidney Diseases[12]

Systematic review RCTs in which antiplatelet agents were compared to placebo or no treatment to evaluate vascular access function in adult ESKD patients with existing vascular access or scheduled to undergo creation of vascular access

Comparatorplacebo or no treatment

Interventionantiplatelet therapy

9/11 Yes Yes Yes Yes NoCommentno list of excluded studies

Characteristics Included Studies Given?

Scientific Quality Studies Assessed?

Scientific Quality Studies Used?

Methods To Combine Findings Correct?

Likelihood Publication Bias Assessed?

Conflict Of Interest Stated?

N Citations Evaluated For Inclusion?

N Full Papers Evaluated For Inclusion?

N Studies Enrolled?


Yes Yes Yes Yes No Yes 1020 171 21 4790

Outcome Outcome Defined As

Time Point Of Outcome

Outcome Value Comparator Group




Effect Measure




P-Value Evidence Certainty

Judgement Comments

Access failure

Early Vascular Access Failure after AV fistula creation

8 weeks 116 61 552 553 Relative Risk

0.43 0.26 0.73 0.002 Yes, but these are overall results without differentiation for type, start and duration of antiplatelet therapy

Maturation failure

Ability to use AV fistula for dialysis with 2 needles and maintain a blood flow rate > 300 mL/min during 8 of 12 dialysis sessions

Up to 6 weeks

92 55 448 454 Relative Risk

0.47 0.19 1.16 0.1 No, small number of participants in only 2 studies, one of which published in 1985

Limitations: - small number of participants ins 2 studies - high heterogeneity

Primary assisted patency

Need for intervention to attain patency or assist maturation (surgical revision or angioplasty) of AV fistula

6 weeks 10 7 431 435 Relative Risk

0.69 0.26 1.83 0.5 No, only from one study

of 226

Major bleeding

Includes gastrointestinal bleeding but is not otherwise described by original studies

up to six weeks


1.15 0.37 3.61 0.8 No, only reported in three studies, two of them published in 1974 and 1985


Study Design Patient Characteristics Inclusion Comparator - Type

Intervention - Type

Amstar Score






ViecelliOmega-3 Polyunsaturated Fatty Acid Supplementation to Prevent Arteriovenous Fistula and Graft Failure: A Systematic Review and Meta-analysis of Randomized Controlled Trials2018American Journal of Kidney Diseases[15]

Systematic review

RCTs in which omega-3 polyunsaturated fatty acids were compared to placebo or no treatment to evaluate vascular access function in for prevention of vascular access failure in adult receiving or planning to receive haemodialysis using an AV fistula, AV graft, or AV shunt in upper or lower limb


Interventionomega-3 polyunsaturated fatty acids

10/11 Yes Yes Yes Yes NoCommentno list of excluded studies









N Studies Enrolled? N Patients Enrolled?

Yes Yes Yes Yes Yes Yes 128 15 6 865


Time Point of the Outcome





Effect Measure





All cause mortality

- 12 to 52 weeks


0.98 0.51 1.86 Not reported

Very low4 RCTs Imprecision: risk estimate includes null effect Directness: data derived from small no. of studies in specific settings that may not be generalizable (treatment duration variable, access different; interventions may differ)

Primary patency failure

First thrombosis

12 to 52 weeks


0.81 0.66 0.98 <0.05 Moderate3 RCTs

of 226

or need for surgical or endovascular intervention to restore patency

Directness: data derived from small no. of studies in different settings that may not be generalizable (treatment duration variable, access type variable)



12 to 52 weeks



Low2 RCTsImprecision: risk estimate includes null effect Directness: data derived from small no. of studies in specific settings that may not be generalizable (treatment duration variable, access different; interventions may differ)

Vascular access abandonment

Not reported 12 to 52 weeks




Bleeding Not otherwise defined

12 to 52 weeks



Very low4 RCTs - not info about severity of bleeding Severe imprecision (2 downgrades): risk estimates include null effect and estimate consistent with both appreciable benefit and harm; Directness: data derived from small no. of studies in specific settings that may not be generalizable (treatment duration variable, dose variable, coadministration of antiplatelet agents variable)

of 226


Study Design Location Patient Characteristics Comparator

Intervention 1

Amstar A Priori Design Provided?




BasharRole of far infra-red therapy in dialysis arterio-venous fistula maturation and survival: systematic review and meta-analysis2014PLOS One[16]

Systematic review

Asia InclusionRCTs and observational studies to examine far-infrared therapy in ESKD patients with AV-fistula. Exclusioncase series and case reports.

Comparatorno treatment

Interventionfar infrared therapy

9/11 No Yes Yes Yes

List Of In- & Excluded Studies? Characteristics Included Studies Given?

Scientific Quality Studies



Conflict of Interest Stated?

Evaluated For Inclusion

N Studies Enrolled?

N Patients Age

Male Sex %

no

Commentonly list of included studies

Yes Assessed?Yes

Used?Yes

Yes Yes Yes N citations1244

N full papers8

4 Enrolled? 666Included Comparator Group?326Included Group 1?340

Age62 years +/-15 years

Male Sex %52%

Outcome Defined As



Outcome Value Intervention Group 1


Analysed Participants Intervention Group 1

Effect Measure




Results Are Reliable?

Primary patency: need for any interventional procedure (surgery or angioplasty) to correct an occlusive or malfunctioning fistula which could not sustain an extracorporeal blood flow < 200 mL/min during hemodialysis

12 months 185 228 299 311 Relative Risk

1.23 1.12 1.35 Yes, but all studies from same centre

Secondary patency: following salvage procedure 12 months 140 160 163 168 Relative Risk

1.11 1.04 1.19

of 226



Comparator - Type

Intervention - Type

Amstar Score






WanEffects of far infrared therapy on arteriovenous fistulas in hemodialysis patients: a meta-analysis2017Renal Failure[17]

Systematic review

RCTs comparing Far-infra red therapy versus no treatment for promoting maturation in adults receiving haemodialysis using an AV fistula


InterventionFar infrared therapy

8/11 No Yes Unclear

Comment-many databases including several Chinese-Complete search strategy not provided

Yes No

Commentno list of excluded studies









N Studies Enrolled?









Effect Measure




P-Value

Certainty evidence

Primary patency

Not otherwise defined

12 months



Low4 RCTs Risk of bias: unclear risk of biasData derived from small no. of studiesDirectness: only 5/21 contributed to meta-analysis. Quality of Chinese language trials difficult to assess.

AV fistula occlusion


12 months



Low5 RCTsRisk of bias: most studies at unclear risk of biasDirectness: only 5/21 contributed to meta-analysis. Quality of Chinese language trials difficult to assess. Definitions unclear

AuthorTitle

Study Design

Patient Characteristics Inclusion

Comparator - Type Amstar Score

A Priori Design

Duplicate Study

Comprehensive Literature

Publication Inclusion

of 226

Publication YearJournal

Intervention - Type Provided? Selection/ Extraction?

Search? Criterion?

SmithA systematic review and meta-analysis of systemic intraoperative anticoagulation during arteriovenous access formation for dialysis2016Journal of Vascular Access[18]

Systematic review

RCTs testing systemic anticoagulation duringaccess formation versus a control group without systemic anticoagulation reporting bleeding complications and access patency

Comparatorplacebo or no treatmentInterventionintraoperative systemic anticoagulation

5/11 No Yes No

Commenttime frame limited to 2015 onwards

Yes










N Studies Enrolled?


NoComment: no list of excluded studies

Yes Yes Yes Unclear Unclear No 445 3 4 411

Evide

Outcome

Defined As






Effect Measure




P-Value

Evidence Certainty Judgement Comments

Access patencyEarly Vascular Access Failure after AV fistula creation

Up to 1.5 months

- - - - Relative Risk

0.64 0.37 1.09 0.1 Low, included studies at high risk of bias and large imprecision of the summary effect estimate

Bleeding

Up to 1.5 months

- - - - Relative Risk

7.18 2.41 21.38 <0.001 Low, included studies at high risk of bias, and total sample size below optimal information size.

Limitations: - small number of participants ins 2 studies - high heterogeneity

of 226


Study Design LocationN Centres

Start/End Of Study (Year)

Patient Characteristics Comparator - Type

Intervention Sequence Adequately Generated?

Allocation Adequately Concealed?

AbacilarOral prostacycline analog and clopidogrel combination provides early maturation and long-term survival after arteriovenous creation: a randomized controlled study2015Indian Journal of Nephrology[23]

Randomised controlled trial

Turkey1

2008 - 2013 Inclusionpeople set to undergo AV fistula creation Exclusionactive or past bleeding; platelet count <75000/µL; coagulopathy; acute ulcer disease; blood pressure >200/115 mmHg; advanced liver disease; history of esophagitis or gastritis; discontinued antiplatelet; pregnancy or lactation; history of myocardial infarction; cerebrovascular accident in previous 12 months

Placebo Type + doseoral clopidogrel 75 mg once daily + oral prostacyclin analogue 200 mg once daily

Duration7-10 days before to one year after surgery

Yes

Commentrandomization stratified according to the medical centre with a permuted block scheme with block size = 4

Yes

Commentsequence was not known to the participant or personnel

Participants Blinded To Treatment?

Personnel Blinded To Treatment?

Outcome Assessment Blinded?

Outcome Measured For All Part?

% Lost To Follow-Up?

All Expected Outcomes Measured?

N Patients AgeMale Sex %

Other Main Characteristics?

Yes

Commentmatching placebo; medication not known to the participant

Yes

Commentmedication not know to personnel

Yes

Commentparticipants of the study team were blinded to the treatment

Yes 0% Yes Evaluated for Inclusion?385Enrolled?96Included Comparator Group?46Included Group 1?50

Age55 years +/-1 year

Male Sex %68%

OutcomeDefined As






Effect Measure

Value 95%CI, P-Value

Evidence Certainty

Maturationblood flow ≥ 300 ml/min, velocity ≥ 70 cm/sec


1.28 1.01 to 1.61

Low. Proportion maturation such defined decreased over time; reported as proportions, denominators are unclear; low external validity due to exclusion criteria

Primary AV fistula failurenew fistula creation

1 year 14 43 4 50 Relative Risk

0.26 0.09 to 0.74

High, but low external validity due to exclusion criteria

Bleeding 1 year Not reported 0 46 50 Not reported

Not reported

Not reported

Low, all patients prone to bleeding were excluded from the study

AuthorTitle

Study Design

LocationN Centres

Start/End of Study

Patient Characteristics Comparator Type

Intervention 1 Intervention 2 Sequence adequately

of 226

Publication YearJournal

(year) generated?

ChenCombined use of heparin and anisodamine reduces the risk of early thrombosis in native arteriovenous fistula2013Vascular[30]


Asia1

2009- 2011

Inclusionstage 4 or 5 CKD patients expected to have hemodialysis within the next six months and a planned lower arm AV fistula

Exclusiontendency for bleeding, active peptic ulcer, platelet count <80 × 109/L), contraindications of anticoagulant therapy, glaucoma, prostatic hyperplasia, nephropathy due to secondary factors

no treatment TypeheparinDose50 IU/kg of heparin intravenous drip in l00 mL of normal salineDurationonce daily, immediately after AV fistula creation and then continued for seven days

Typeheparin plus anisodamineDose50 IU/kg of heparin plus 10 mg of anisodamine intravenous drip in l00 mL of normal salineDurationonce daily, immediately after AV fistula creation and then continued for seven days

no

Commentpatients were randomised by proper order of the number of operation into three groups








N Patients Evaluated For Inclusion?

N Patients Age

Male Sex %


Unclear

Commentnot reported

No

Commentcontrol group received no treatment, at least this group was unblinded

No

Commentcontrol group received no treatment, at least this group was unblinded

Unclear

Commentnot reported

Yes 0

Commentno information provided

No Not reported Enrolled: 180Comparator Group: 60Group 1: 60Group 2: 60

55 years +/- 16 years

54%

All secondary nephropathies excluded

OutcomeDefined As








Effect Measure

P-Value Evidence certainty

Primary patencyobvious thrill, AV fistula diameter 5-7mm, blood flow rate 100-300cm/sec, no need for re-intervention

83% 87% 97% 60 60 60 Success rate, not further explained

Low, unclear how success rate was calculated

Access survivalthrombosis

4 weeks 10 8 2 60 60 60 Chi-square test

0,047 for heparin compared to heparin/anisodamine; 0,015 for control group compared to heparin/anisodamine;0,399 for control group compared to heparin

Low, inadequate randomization, unblinded, patients with secondary nephropathies excluded

AuthorTitlePublication Year

Study Design

LocationN Centres


Patient Characteristics

Comparator - Type

Intervention Sequence Adequately Generated?


of 226

JournalBleyerThe effect of topically applied Vonapanitase on fistula outcomes: results of the PATENCY-1 Trial2017American Journal of Kidney Diseases(Abstract)[24]


United States of America31

Not reported

Inclusionpeople set to undergo AV fistula creation

Exclusionnot reported

Placebo Type + dosevonapanitaseafter AV fistula creation dropwise on exposed artery, anastmosis and vein

Duration10 min, immediately after fistula creation

Unclear

CommentAbstractrandomised

Unclear

CommentNot reported









Yes

Commentmatching placebo

Yes

Commentdouble-blind

Unclear

CommentNot reported

Yes 0% No, adverse events not reported

Evaluated for Inclusion?Not reportedEnrolled? 313Included Comparator Group?Not reportedIncluded Group 1?Not reported

2:1 randomisation

AgeNot reported

Male Sex %Not reported

Not reported

OutcomeDefined As






Effect Measure


Evidence Certainty

Primary unassisted patencytime to first thrombosis or intervention to restore flow

1 year 31% 42% Not reported Not reported Not reported

Not reported

0.25 Unclear

Secondary patencytime to abandonment


Not reported

0.048 Unclear

of 226


Study Design Location Number Of Centres

Patient Characteristics Comparator Intervention Sequence Adequately Generated?


WasseVery high-dose cholecalciferol and arteriovenous fistula maturation in ESKD: a randomised, double-blind, placebo-controlled pilot study2014Journal of Vascular Access[25]


North America

1 InclusionESKD patients on chronic hemodialysis scheduled to receive an AV fistula within 4 weeksExclusionnot reported

TypeplaceboDoseonce weekly

Typeoral vitamin D3 (cholecalciferol)Dose200,000 IU, once weeklyDuration3 weeks

Yes

Commentsubjects were randomly assigned by using block randomization

Yes

Commentrandom assignment was conducted by an independent clinical trial pharmacist





% Lost To Follow-Up? All Expected Outcomes Measured?

N Patients Age

Male Sex %


Yes

CommentStudy personnel and subjects were blinded to the treatment arm. Treatment allocation was not revealed until study completion.

Yes

CommentStudy personnel and subjects were blinded to the treatment arm. Treatment allocation was not revealed until study completion.

Unclear

Commentno information

Yes 15%

Reasons1 patient was lost to follow-up; 3 patients died; 4 patients never received a permanent vascular access

Yes Evaluated for Inclusion?107Enrolled?52Actually Started?44Included Comparator Group?24Included Group 1?20


68%

91 % black

OutcomeDefined As






Effect Measure

P-Value

Evidence Certainty

Maturationability to cannulate with two large-bore needles at ≥6 consecutive dialysis sessions and achievement of blood flow >300 mL/min

6 months following AV fistula/AV graft creation

50% 41% Not reported Not reported Fisher's exact test

0.7 Low, single centre and pilot study with small number of participants

of 226


Study Design

LocationN Centres


Patient Characteristics Comparator Intervention Sequence Adequately Generated?


FieldRandomised clinical trial of the use of glyceryl trinitrate patches to aid arteriovenous fistula maturation2016Nephrology Dialysis Transplantation(Abstract)[26]


United Kingdom1

2013-2015

Inclusionage >18 years; set to undergo radio-cephalic or brachiocephalic AV fistula

Exclusioncomplex vascular access procedures; cardiovascular dysfunction; anaemia; migraine; silfadenil or other nitrate use; nitrate allergy; closed-angle glaucoma; chronically raised intracranial pressure; hypothyroid disease; pregnancy; prisoners; veins smaller than 2 mm.

Type + doseplacebo patch

Durationend of surgery until 24 h after surgery

Type + doseglyceryl trinitrate patch 5 cm proximal to anastomosis

Durationend of surgery until 24 h after surgery

Yes

Commentrandomisation using varying block length randomisation - via telephone

Yes

Commentrandomisation via telephone









Yes

Commentmatching placebo patch - all patients were blinded to randomisation

Yes

Commentall members of the study were blinded to randomisation

Unclear

CommentNot reported

No

Comment16.5% not included in outcome measurement due to abandonment of surgery of lost to follow-up

10% Yes Evaluated for Inclusion?533Enrolled?200Included Comparator Group?99Included Group 1?101

Age60 years ± 15 years

Male Sex %60%

All fistulas were constructed end-of-vein to side-of-artery under local anaesthetic

OutcomeDefined As






Effect Measure

Value 95%CI, P-Value Evidence Certainty

Primary patency 6 weeks 77% 72% 81 86 Relative Risk

1.2 0.7 to 2.2; 0.60 Moderate

Change in venous diameterchange in mm

6 weeks 2.3 ± 1.9 mm 22 ± 1.8 mm 81 86 Mean Difference

-0.1 -0.7 to -0.5; 0.70 Moderate

of 226


Study Design

LocationN Centres


Patient Characteristics Comparator Intervention 1Intervention 2

Sequence Adequately Generated?


RongrongThe effects of iontophoretic injections of Salvia miltiorrhiza on the maturation of the arteriovenous fistula: a randomised, controlled trial2016Alternative therapies in health and medicine[27]


China1

Not reported Inclusionage 18-85 years; had undergone radio-cephalic AV fistula creation

Exclusioninjury or skin disease on fistal arm; heart disease requiring anticoagulatory drugs; coagulation disorders; peripheral artery disease; smoking.

Type + dosestandard care

Type + doseiontophoretic tradermal injection of Salvia milthiorrhiza

Durationstarting day of surgery; 30 minutes; twice daily for 1 month

Yes

Commentpermuted block randomisation was used

Yes

Commentinvestigators were blinded to block size









No

Commentcomparison with standard care

No

Commentdue to nature intervention and comparison with standard care

Unclear

CommentNot reported

Yes 0% No Evaluated for Inclusion?40Enrolled?40Included Comparator Group?20Included Group 1?20

Age control55 years ± 12 yearsAge intervention61 years ± 12 yearsMale Sex %60%

All fistulas were constructed end-of-vein to side-of-artery

OutcomeDefined As





Analysed Participants Intervention Group 1Group2

Effect Measure


Evidence Certainty

Maturation2 needles, blood flow ≥300 mL/min ≥6 dialysis sessions in 1 month, within 6 months of creation

1 month 10 17 20 20 Relative Risk

1.7 1.06 to 2.73 Low; small sample size below expected minimal information size; wide confidence interval; time-point early for measuring maturation; Problems with generalizability and applicability. Maturation definition in methods state outcome measurement during 6 months; reported only for one month.

Adverse reactionsnot otherwise defined


Not estimable

Not estimable

No detailed report of what was assessed. Small sample size.

of 226


Study Design

LocationN Centres


Patient Characteristics Comparator Intervention 1Intervention 2

Sequence Adequately Generated?


VoorzaatImprovement of radio-cephalic fistula maturation: rationale and design of the Liposomal Prednisolone to Improve Hemodialysis Fistula Maturation (LIPMAT) study - a randomised controlled trial2017Journal of Vascular Access[28]


The Netherlandsmulticentre

2016-? Inclusionage not reported; having received radio-cephalic AV fistula

Exclusionipsilateral central venous catheter, current malignancy, latent or active infections with tuberculosis or hepatitis B and C, uncontrolled diabetes mellitus and known contraindications to glucocorticoids; use of systemic glucocorticoids, immunosuppressant medication or nonsteroidal anti-inflammatory drugs

Type + doseintravenous 500 mL saline

Durationday 1 and day 14 after surgery

Type + doseintravenous liposomal prednisolone 150 mg in 500 mL saline

Durationday 1 and day 14 after surgery

Yes

Commentpermuted block randomisation was used

Yes

Commentinvestigators were blinded to block size

Participants Blinded To Treatment? Personnel Blinded To Treatment? Outcome Assessment Blinded?




N Patients

AgeMale Sex %


Yes

CommentBlinding is achieved through a fully opaque IV set equipped with an air filter permeable to the 110 nm liposomes

Yes

CommentBlinding is achieved through a fully opaque IV set equipped with an air filter permeable to the 110 nm liposomes

Unclear

CommentNot reported

Protocol Protocol Protocol Protocol Age control55 years ± 12 yearsAge intervention61 years ± 12 yearsMale Sex %60%

All fistulas were constructed end-of-vein to side-of-artery

of 226

Chapter 2. Surgical and endovascular interventions for promoting AV access maturationAuthorTitlePublication YearJournal



AMSTAR score

A priori design provided?

Duplicate study selection/ extraction?

Comprehen-sive literature search?

Publication inclusion criterion?


IsmailRegional versus local anesthesia for arteriovenous fistula creation in end-stage renal disease: a systematicreview and meta-analysis2017Journal of Vascular Access[35]

Systematic review Randomised trials and observational studies comparing efficacy/safety of regional versus local anaesthesiain AV fistula surgical construction

Comparatorlocal anaesthesia

Interventionregional anaesthesia

8/11 No Yes Yes Yes NoCommentno list of excluded studies









N studies enrolled?


Yes Yes Yes Yes NA no 97 16 7 870







Effect Measure




P-Value

Evidence Certainty

Fistula Thrombosis

- not specified


0.21 0.03 1.27 0.09 Very low (2 RCTs)

Blood flow - after discharge

44.8+/-13.8 69.8+/-7.9 39 39 Mean Difference

25.08 19.40 30.86 <0.001 Very low (2 RCTs)

Primary patency rate

intervention free access survival

not specified


1.22 1.08 1.37 0.001 Very low (3 RCTs)

Primary fistula failure

- not specified


0.81 0.47 1.40 0.46 Very low (3 studies - 2 RCTs)

of 226


Study Design

LocationNumber of centres

Start/End of Study (year)

Patient Characteristics Comparator Intervention

AitkenEffect of regional versus local anaesthesia on outcome after arteriovenous fistula creation: a randomised controlled trial.2016Lancet[37]


Europe1

2013-2015 Inclusion: English-speaking adults aged 18 to 85 competent to give consent and scheduled for primary AV fistula formation at either the radial or brachial arteryExclusion: allergy to local anaesthetic, coagulopathy, infection at the anaesthetic or surgical site, patient preference for general or alternative anaesthesia, significant peripheral neuropathy or neurologic disorder affecting the upper extremity, pregnancy, previous AV fistula creation on the ipsilateral arm, known cephalic vein occlusion, central vein stenosis, brachial or radial artery stenosis and vein or artery less than 1.8 mm, as measured by ultrasound.

Type: infiltration of local anaesthetic into the surgical site by the operating surgeon Dose: combination of 0.5% L-bupivacaine prior to incision and 1% lignocaine topically after the wound is opened

Type: ultrasound guided brachial plexus blockDose: 1:1 mixture of 0.5% L-bupivacaine and 1.5% lignocaine with adrenaline

Sequence adequately generated?

Allocation adequately concealed?

Participants blinded to treatment?

Personnel blinded to treatment?

Outcome assessment blinded?

Outcome measured for all part?

% Lost to follow-up?

All expected outcomes measured?

Additional Sources of Bias

N Patients Age Male Sex %

Yes

Comment: We generated the randomisation sequence (in blocks of eight) using a web-based computerrandom-sequence generator

Yes

Comment: sequentially numbered,opaque, sealed envelopes

No

Comment: Due to the nature of the intervention, it was not possible to blind the patient, surgeon, anaesthetist, or study team involved at the time of surgery.

No Yes

Comment: The vascular access nursespecialist who assessed the primary endpoint wasindependent and blinded to the randomisation.

Yes 0% Yes No Evaluated for Inclusion: 163Enrolled: 126Actually started: 125Included Comparator Group: 63Included Group 1: 63

61years +/- 15years

63%


Time Point of the Out-come




AnalysedParticipants Intervention Group

Effect Mea-sure

Value Effect Mea-sure



P-Value

Evid-ence Certain-ty

Judgement Comments

Primary patency

Presence ofa thrill or bruit in the absence of any additionalintervention to re-establish function

3 months 62% 84% 63 63 Odds Ratio

3.3 1.4 7.6 0.005 High Better patency for brachial plexus anaesthesia remains significant in subgroup analysis comparing radio-cephalic vs brachiocephalic fistulas

of 226

Immediate patency

Patency at time of discharge from hospital

Dis-charge after surgery

73% 92% 63 63 Odds Ratio

4.3 1.5 12.5 0.005 High Better patency for brachial plexus anaesthesia remains significant in subgroup analysis comparing radio-cephalic vs brachiocephalic fistulas

Functional patency

Clinically as used for dialysis or in predialysis patients deemed suitable for cannulation by the vascular access nurse specialist and by ultrasound (>6 mm diameter, <6 mm from skin surface, fl ow rate >600 ml/min)

3 months 29% 41% 63 63 Odds Ratio

1.8 0.8 3.7 0.15 High No difference between brachial plexus anaesthesia and local anaesthesia in brachio-cephalic fistulas, better patency for brachial plexus block in radio-cephalic fistulas


Study Design



Patient Characteristics Comparator Intervention 1 Sequence adequately generated?



MeenaUltrasound-guided supraclavicular brachial plexus anaesthesia improves arteriovenous fistula flow characteristics in end-stage renal disease patients2015Southern African Journal of Anaesthesia and Analgesia[38]


India1

2011-2012

Inclusion: Patients between 18 and 60 years of age of both genders scheduled for hemodialysisExclusion: local infection, allergy to local anaesthetics, clinically significant coagulopathy, anatomical variation at the intended site of brachial plexus injection, body mass index more than 35, revision of previously blocked AV fistula andfailed surgical procedure

Type: anaesthesia by local infiltration

Type: anaesthesia bysupraclavicular brachial plexus block

Yes

Comment: quote "Patients were randomised into two groups by computer-generatedrandomisation number using block randomisation design with a block size of six"

Unclear

Comment: not stated

Yes

Comment: single blind

of 226






N Patients Age

Male Sex %


Unclear

Comment: not reported

Yes 0 Yes No Evaluated for Inclusion: 60Enrolled:60Actually started: 60Included Comparator Group: 30Included Group 1: 30

~45 years

77%

Standardised surgical technique was used for creation of all AV fistula

Outcome Time Point of the Outcome





Effect Measure P-Value Evidence certainty

Judgement Comments

Primary patency Six weeks 90% 100% 30 30 Proportion 0.237 High AV fistula diameter

Six weeks Not stated Not stated 30 30 Outcome results narratively provided "significantly more in the brachial plexus analgesia group versus local infiltrationgroup"

p<0.05 Low, there are serious limitations

No actual numbers provided, no further details on outcome assessment

Peak systolic velocity




Mid-diastolic velocity




Blood flow Six weeks Not stated Not stated 30 30 Outcome results narratively provided "significantly more in the brachial plexus analgesia group versus local infiltrationgroup"

p<0.05 No, there are serious limitations



Study Design LocationNumber of centres



Comparator Intervention 1




ShoshiashviliInfluence of type of anesthesia on hemodynamic parameters and outcome of dialysis arteriovenous fistula operations2015Georgian Medical News[39]


Georgia1

Not stated Inclusion: not statedExclusion: not stated

Type: regional anaesthesia by vertical infraclavicular block

Type: local anaesthesia group

Unclear

Comment: not stated

Unclear

Comment: not stated

Unclear

Comment: not stated

of 226


Outcome assessment blinded? Outcome measured for all part?




no Unclear

Comment: not stated

unclear 0 unclear Evaluated for Inclusion: not statedEnrolled:103Actually started:103Included Comparator Group: 49Included Group 1: 54

~60 years

Not stated






Effect Measure P-Value Evidence Certainty

Judgement Comments

Number of dialysis punctures between the groups

Not stated

Not stated Not stated 49 54 Outcome results narratively provided: "no statistically significant difference detected"

Not provided

Low, there are serious limitations

No actual numbers provided, elements of study design and conduct not reported, functionality of access not reported, subjective outcome with non-standard definition

Rate of thrombosis Not stated


Not provided


No statistical significance and actual numbers provided, elements of study design and conduct not reported, functionality of access not reported, subjective outcome with non-standard definition

Rate of immaturation

Not stated


Not provided



Maturation time Not stated


Not provided



Time to first puncture

Not stated


Not provided



of 226




Patient Characteristics Comparator Intervention 1

SahinUltrasound-guided infraclavicular brachial plexus block enhances postoperative blood flow in arteriovenous fistulas2011Journal of Vascular Surgery[40]


Europe1

Not reported Inclusion: age 18 to 60 years, primary AV fistula access surgery before hemodialysisExclusion: radial artery diameter > 3 mm, cephalic vein diameter <2.5 mm or >4 mm, previous antecubital fistulas, cephalic vein occlusion, ipsilateral central vein stenosis, ipsilateral brachial or radial artery stenosis or calcification, coagulopathy

Type: local infiltration anaesthesiaDose: 10-mL volume of 2% lidocaine

Type: Ultrasound-guided infraclavicular brachial plexus blockDose: 10 mL of 0.5% levobupivacaine and 10 mL of 2% lidocaine







N Patients Age

Male Sex %

Unclear


Unclear


No No Yes

Comment: Postoperative evaluation ... By another blinded surgeon.

Yes Yes Evaluated for Inclusion: not reportedEnrolled: 60Actually started: 60Included Comparator Group: 30Included Group 1: 30

45 years+/- 12 years

43%

OutcomeDefined As






Effect Measure

P-Value

Evidence Certainty

Judgement Comments

Primary patencyintervention-free access survival

8 weeks 25 28 30 30 Not reported

0.23 Low, there are serious limitations

Small sample size, contradicting results reported: access failures occurred in both groups, but blood flow reported for all 30 patients in each group

Assisted primary patency thrombosis-free access survival


0.99 Low, there are serious limitations


Blood flowAV fistula blood flow assessed by duplex ultrasound in ml/min

8 weeks 405 +/- 76 680 +/- 97 Unclear Unclear Mean difference



of 226





Comparator Intervention Sequence adequately generated?


ThomsenAdjuvant intravenous sympathetic block with guanethidine in construction of arteriovenous fistulas for blood access1983Acta Chirurgica Scandinavica[41]


Europe1

Not reported

Inclusion: non-diabetic, uraemic, pre-dialysis patients referred for creation of AV fistula Exclusion: not reported

Type: usual care

Type: Regional sympathetic block by i.v. Injection of GuanethidineDose: 0.25mg/kg bwDuration: single dose into exsanguinated arm, arterial occlusion maintained for 20 min with an inflatable tourniquet 24 hours before surgery

Unclear


Unclear








N Patients Age

Male Sex %


No

Comment: Guanethidine was only injected in interventional group within 24 hours before surgery

No No No

Comment: number of patients in whom outcome was assessed not reported for most outcomes

Unclear


No Evaluated for Inclusion: not reportedEnrolled: not reportedActually started: 34Included Comparator Group: 17Included Group 1: 17

55 years (17 years -22 years)

68%

All pre-dialysis

OutcomeDefined As







Judgement Comments

Early fistula failurepatency loss within 2 weeks after surgery


High But very old study

Blood flowblood flow through vein after fistula creation

At surgery after completion of anastomosis

75 ± 40 ml/min 108 ± 47 ml/min 8 Not reported Not reported

Low Old study, unclear in how many patients’ outcome assessed

of 226


Study Design



Patient Characteristics Comparator Intervention Sequence adequately generated?



Yildirim Does pre-emptive stellate ganglion blockage increase the patency of radio-cephalic arteriovenous fistula?2006 Scandinavian Cardiovascular Journal[42]


Europe1

End of study 2006

Inclusion: Patients having a radio-cephalic fistula not further specifiedExclusion: previous antecubital fistulas, cephalic vein occlusion, ipsilateral central vein stenosis, ipsilateral brachial or radial artery stenosis

Type: usual care

Type: stellate ganglion blockage 1 hour before surgeryDose: 10 ml Ropivacaine 1%Duration: continued for 7 days following surgery

Yes

Comment: randomization by drawing lots

Unclear


No

Comment: no stellate blockage in control group

Personnel blinded to treatment? Outcome assessment blinded?




N Patients Age

Male Sex %

No

Comment: surgeon was blinded, but stellate ganglion blockage leads to increased skin temperature

No

Comment: only surgeon was reported to be blinded

Unclear

Comment: number of patients assessed for outcomes not reported

Unclear



57 years ± 10 years

76%

OutcomeDefined As






Effect Measure

P-Value

Evidence Certainty

Judgement Comments

Thrombosisnot further specified

24 hours after surgery

8 2 Not reported Not reported Not reported

0.034 Low Number of patients in whom outcome was assessed is not reported

Time until maturationdecided by surgeon and nephrologists based on thrill characteristics and diameter of arm veins

Not reported 77.1 ± 10.5 41.4 ± 6.8 Not reported Not reported Not reported

<0.001 Low Number of patients in whom outcome was assessed is not reported, various definitions of maturation throughout manuscript

Adequate vascular accesssuccessful cannulation of superficial arm vein for hemodialysis without excessive effort for cannulation (maximum two or three times).

Not reported 12 19 Not reported Not reported Not reported

0.041 Low Number of patients in whom outcome was assessed is not reported

of 226


Study Design



AMSTAR score



Compre-hensive literature search?



BasharEnd-To-Side versus Side-To-Side Anastomosis in Upper Limb Arteriovenous Fistula for Dialysis Access: A Systematic Review and a Meta-Analysis2018Annals of Vascular Surgery[36]

Systematic review

Randomised controlled trials and observational studies that compared the end-to-side versus the side to-side anastomosis techniques in creating an upper limb AV fistula

Comparatorside to-side

Interventionend-to-side

7/11 no yes Yes Yes NoCommentno list of excluded studies









N studies enrolled?


Yes Yes Yes Yes Not available no 628 13 7 986







Effect Measure




P-Value

Evidence Certainty

Patency rate - 3 months 62 67 58 58 Relative Risk

0.92 0.8 1.07 0.28 Very low 2 studies (1 RCT)


0.98 0.86 1.13 0.82 Very low 5 studies (2 RCTs)


1.06 0.87 1.30 0.54 Very low 3 studies (no RCTs)


1.07 0.96 1.18 0.21 Very low 2 studies (no RCTs)






WedgwoodA prospective study of end-to-side vs. side-to-side arteriovenous fistulas for haemodialysis1984British Journal of Surgery[43]

Europe1

1981-1983 Inclusion: All patients requiring primary vascular access surgery for haemodialysis Exclusion: not reported

Type: End of vein to side of artery anastomosis

Type: Side to side anastomosis

No

Comment: not reported and in case of surgical difficulties allocation in alternative group by surgeon

No

Comment: not reported and surgeon allocation to alternative group in case of surgical difficulties

of 226








Unclear


No Unclear


Unclear


Unclear


Unclear Evaluated for Inclusion: not reportedEnrolled: not reportedActually started: 71Included Comparator Group: 39Included Group 1: 32

Not reported

68%

OutcomeDefined As







Judgement Comments

Access patencyfistula used for dialysis

3 months 87% 87% 38 30 Not reported High But very old study

Access patencyfistula used for dialysis

9 months 79% 79% 28 24 Not reported High But very old study

Primary failurethrombosis

Not reported 10% 9% 39 32 Not reported High But very old study






MozaffarComparison of efficacy of side to side versus end to side arteriovenous fistulae formation in chronic renal failure as a permanent hemodialysis access2013Nephro-Urology Monthly [44]


Middle-East1

2010-2012 Inclusion: patients referred for vascular access replacement, without injection or blood sampling during the last couple of weeks, systolic blood pressure > 100 mmHg

Type: side to side anastomosis

Type: end to side anastomosis

Unclear


Unclear






% Lost to follow-up? All expected outcomes measured?

N Patients Age

Male Sex %Unclear


No Unclear


Unclear


Unclear


No Evaluated for Inclusion: not reported (study cohort is a simple random sample of all patients who were referred)Enrolled: 60Actually started: 60Included Comparator Group: 30Included Group 1: 30

Not reported

not reported

of 226

OutcomeDefined As






Effect Measure


Judgement Comments

Access survivalineffective fistula not further defined

6 months 6 5 Not reported Mot reported Not reported

Not reported Low Extremely poorly designed, performed and reported study.


Study Design




Comparator Intervention 1 Sequence adequately generated?

KhanComparative study of efficacy of end-to-side with side-to-side arteriovenous fistula in patients on hemodialysis2015Pakistan Journal of Medical and Health Sciences[45]


Pakistan1

2012-2013 Inclusion: not reportedExclusion: not reported

End-to-side technique for AV fistula formation

Side-to-side technique for AV fistula formation

Unclear

Comment: randomised controlled trial not otherwise specified








N Patients Age

Male Sex %

Unclear


Yes

Comment: single blind

Unclear Unclear Unclear 0 Unclear Evaluated for inclusion: unclearEnrolled: 336Actually started: 168Included comparator group: 168

40 years +/-0.5 years71%






Effect Measure

Value 95% Confidence interval


Efficacynot otherwise defined


1.05 0.97 to 1.14 0.25 Low. Small study, high risk of bias, incomplete reporting of methodology and outcomes.

Bleedingnot otherwise defined


0.43 0.11 to 1.63 Not reported

Low. Small study, high risk of bias, incomplete reporting of methodology and outcomes.

of 226







SchildPreliminary prospective randomised experience with vascular clips in the creation of arteriovenous fistulae for hemodialysis1999American Journal of Surgery[46]


North America1

1995-1996 Inclusion: not reported

Exclusion: not reported

Type: sutured anastomosis

Type: anastomosis with vascular clipping system

Yes

Comment: Randomization generated by computer programme

Yes

Comment: ...sealed in a randomization notebook controlled by an individual not associated with study patients. Sequential assignments were given as patients accrued and only after eligible patients had consented to participate.







Additional Sources of Bias N Patients Age

Male Sex %

Unclear


No

Comment: surgical intervention

Unclear


Yes 0.01 %

Reasons for lost to follow-up: not reported

No Within treatment arms patients were divided to fistula, graft, or graft revision depending on functional state of patient, anticipated time to initiate dialysis, flow and anatomic characteristics.

Evaluated for Inclusion: not reportedEnrolled: not reportedActually started: 96Included Comparator Group: 49Included Group 1: 46

50 years

61%

OutcomeDefined As






Effect Measure


Judgement Comments

Primary patencySociety for Vascular Surgery/International Society for Cardiovascular Surgery standards

24 months 61% 71% 20 22 Log rank test 0.66 Low Access type was clinical decision and not randomised, small study

Secondary patencySociety for Vascular Surgery/International Society for Cardiovascular Surgery standards

24 months 61% 91% 20 22 Log rank test 0.05 Low Access type was clinical decision and not randomised, small study

of 226


Study Design






WalkerU Clips for arteriovenous anastomosis: a pilot, randomised study2012ANZ journal of surgery[47]


Australia/New Zealand3

Not reported Inclusion: arterial inflow vessel diameter ≥ 2 mm and outflow vein diameter ≥ 3 mm Exclusion: not reported

Type: anastomosis with either a conventional continuous 6/0 Prolene suture

Type: anastomosis with interrupted U Clips

Yes

Comment: using an electronic random number generator

Yes

Comment: procedures were performed at three different sites, each with their own selection of randomization envelopes






N Patients Age

Male Sex %

OutcomeDefined As

Unclear


No

Comment: surgical procedure, surgeon not blinded

Yes

Comment: The operation note did not include the words Prolene anastomosis or U clipped anastomosis. In this way, patients, nephrologists and access nurses were blinded.

Unclear

Comment: number of patients assessed for outcome not reported

Not reported No Evaluated for Inclusion: not reportedEnrolled: 31Actually started: 31Included Comparator Group: 19Included Group 1: 12


68%

Maturation (3 of 4 successful dialysis sessions)three successful dialysis treatments






Effect Measure





Judgement Comments

3 months 42% 58% Not reported Not reported Chi squared test, odds ratio

0.52 0.09 2.82 0.38 High But small sample size, pilot study, number of patients assessed for outcome and lost to follow up not reported


Study Design



Patient Characteristics Comparator

Intervention Sequence adequately generated?


ZeebregtsRandomised clinical trial of continuous sutures or non-penetrating clips for radio-cephalic arteriovenous fistula 2004British Journal of Surgery[48]


Europe1

2000-2003 Inclusion: radial artery diameter ≥ 2 mm, patent palmar arch, cephalic vein diameter ≥ 2,3 mm, adequate venous outflow tract on duplex imaging

Type: end-to-side- anastomosis with 6.0 suture

Type: end-to-side- anastomosis with medium-sized clips

Unclear


Yes

Comment: randomization was done during surgery using a sealed envelope opened by the anaesthetist

of 226




Outcome measured for all part? % Lost to follow-up?



Unclear


No Unclear


Unclear

Comment: number of patients assessed not reported

0 Yes Evaluated for Inclusion: not reportedEnrolled: 98Actually started: not reportedIncluded Comparator Group: not reportedIncluded Group 1: not reported


69%

OutcomeDefined As






Effect Measure

P-Value

Evidence Certainty

Judgement Comments


6 months 61% 69% Unclear Unclear Kaplan Meier statistics

0,39 No, there are serious limitations

Number of patients in each group and assessed for outcomes not reported, unit of analysis is number of fistulas created rather than number of patients, but tests for independent samples nevertheless used







BeigiThe effect of ligation of the distal vein in snuff-box arteriovenous fistula2009Saudi Journal of Kidney Diseases & Transplantation[49]


Middle-East1

2005-2006

Inclusion: not reported, chronic renal failure patients who receive a snuff-box fistulaExclusion: not reported

Type: no vein ligation after anastomosis

Type: Ligation of distal vein after side to side anastomosis

Unclear


Unclear


Unclear








N Patients Age

Male Sex %


OutcomeDefined As

No Unclear


Unclear

Comment: number of patients assessed not reported

Not reported

Yes Vein assessments very subjective. Number of patients assessed in each group unclear

Evaluated for inclusion: not reportedenrolled: 60actually started: 60included comparator group: not reportedincluded group 1: not reported

51 years +/-17 years

50%

Snuffbox fistula Access survivalpresence of thrill

of 226






Effect Measure

P-Value Results are reliable?

Judgement Comments

90 days 83.4% 90% Not reported Not reported Proportion >0.05 No, there are serious limitations

Small sample size, elements of study design and conduct not reported, number of patients assessed for outcome not reported, functionality of access not reported, subjective outcome with non-standard definition






LaskarAnastomosis of small caliber vessels. Comparison between continuous or interrupted suture1988Presse medicale[50]


Europe1

End of study before 1988

Inclusion: non diabetic for their first fistula with vessels < 2mmExclusion: not reported

Type: continuous suture

Type: interrupted suture

Yes

Comment: drawn by lot

unclear








N Patients Age

Male Sex %


Unclear


No Unclear


Unclear


Unclear



Not reported

65%

Not reported

OutcomeDefined As







Judgement Comments

Access survivalpatency

Up to 2 years 87% 83% Not reported Not reported Not significant

No Small, old study with many details not reported

of 226


Study Design



Patient Characteristics Comparator Intervention 1 Sequence adequately generated?



KakkosRandomised controlled trial comparing primary and staged basilic vein transposition2015Frontiers in Surgery[51]


Greece1

2010-2013 Inclusion: patients between 18 and 90 years of age of both genders, with CKDalready on hemodialysis or with anticipated hemodialysisExclusion: unsuitable or already enlarged basilic vein

Type: staged transposed brachio-basilic vein fistula

Type: primary transposed brachio-basilic vein fistula

Unclear

Comment: not stated

Yes

Comment: "Randomization was performed using sequentially numbered sealed opaque envelopes stratified by vein size"

No

Comment: open label (blinding not possible)







N Patients Age

Male Sex %


No Unclear

comment: not reported

Yes 0 Yes No Evaluated for Inclusion: 17Enrolled:17Actually started: 17Included Comparator Group: 7Included Group 1: 9

60,5 years (50,5 years -67 years)

41%

Patients with unsuitable cephalic veins or exhausted cephalicvein






Effect Measure

P-Value

Results are reliable? Judgement Comments

Transposed brachiobasilic fistula maturation

10 weeks 100% 33% 7 9 Proportion <0.01 No, there are serious limitations

small sample size

primary patency 1 year 57% 44% 7 9 Proportion 0.76 No, there are serious limitations

small sample size

secondary patency 1 year 86% 44% 7 9 Proportion 0.09 No, there are serious limitations

small sample size

Non-thrombotic post-operative and transposed brachio-basilic vein fistula-related complications

Not reported

43% 56% 7 9 Proportion 1 No, there are serious limitations

small sample size

of 226


Study Design





VerouxPrimary balloon angioplasty of small (?2 mm) cephalic veins improves primary patency of arteriovenous fistulae and decreases reintervention rates2013Journal of vascular surgery[52]


United States of America, Europe2

2009-2011 Inclusion: radial artery with normal duplex ultrasound parameters and cephalic vein ≤ 2 mm in diameterExclusion: segmental cephalic vein occlusions, brachial and radial artery extensive disease that precluded the creation of a distal autogenous AV fistula

Type: Hydrostatic dilatation of cephalic vein with ligation of collateral veinsDose: high-pressure sterile solution through a syringe 20 mL and 16-gauge plastic cannula

Type: Balloon angioplasty of cephalic veinDose: non-compliant balloon, 4 mm x 150 mm, 12 atmospheres of pressure, 60 seconds, from elbow to level of the anastomosis

Unclear


Unclear


Participants blinded to treatment? Personnel blinded to treatment?





N Patients Age

Male Sex %


NoComment: One patient scheduled for primary balloon angioplasty of the cephalic vein refused the procedure and underwent a standard hydrostatic dilation of the vein.

No Unclear Unclear Reasons for lost to follow-up: not reported

no Evaluated for Inclusion: 130Enrolled: 40Actually started: 40Included Comparator Group: 21Included Group 1: 19


73%

Single surgeon practice

OutcomeDefined As






Effect Measure


Judgement Comments


6 months 57% 95% 14 14 Not reported

Not reported

No, there are limitations

small sample size, analysis by kaplan meier curves only, indications for interventions to prevent loss of access (and loos primary patency) not reported and blinding for decision makers unclear

Reintervention 6 months 43% 5% Not reported Not reported Not reported

0.02 No, there are limitations

small sample size, number of patients assessed not reported, test used unclear

Maturationnumber of patients with an AV fistula useful for hemodialysis access

Not reported 90% 100% 21 19 Not reported

0.5 No, there are limitations

small sample size, maturation not precisely defined.

of 226

Chapter 3. Surgical and endovascular interventions for non-maturing fistulas

Endovascular interventions for non-maturing fistulas

AuthorYear of publication

N participants Target lesion/Technique

ClinicalSuccess (%)

1-yearPrimary patency (%)

1-yearSecondary patency (%)

Beathard 1999 63 VO/AVO 82 -- 75Turmel Rodrigues 2001 69 JAS/VO 97 39 79Beathard 2003 100 JAS/VO/AVO 92 -- 68Tordoir 2003 12 JAS/VO 43 -- --Shin 2005 19 JAS 74 61 82Falk 2006 65 JAS/VO 74 64 68Song 2006 22 JAS/VO 95 28 85Asif 2006 41 JAS 93 46 94Nassar 2006 119 JAS/VO/AVO 83 62 94Barone 2007 43 AI/JAS/VO 85 -- --Clark 2007 101 JAS/VO 88 34 72Singh 2008 32 JAS/VO/AVO 78 -- --Ascher 2009 27 BAM 89 -- --Hong 2009 8 JAS 87 56 87McLafferty 2009 23 VO 93 -- --Raynaud 2009 25 AI 91 83 86Turmel Rodrigues 2009 74 AI 98 65 96Natário 2010 30 JAS 97 37 97Miller 2011 140 VO/AVO 79 -- 90Gallagher 2012 45 BAM 83 -- --Han 2013 141 JAS/VO 96 72 83Derderian 2013 30 BAM 55 -- --Ahmed 2014 42 AVO 76 -- --Park 2015 24 VO 96 59 --Jeon 2016 59 JAS/VO 95 71 --Rizvi 2017 54 BAM 56 -- --

of 226

Park 2017 84 JAS/VO/AVO -- -- 94Bavare 2017 12 VO/ stent graft 100 65 72

JAS= Juxta-anastomotic stenosis; AI=arterial inflow; VO=venous outflow; AVO=accessory vein obliteration; BAM=balloon-assisted maturation (ultrasound-guided). Adapted from Tordoir et al. [53]

Surgical interventions for non-maturing fistulas

AuthorYear of publication

N participants Technique

ClinicalSuccess (%)

1-yearPrimary patency

(%)1-year

Secondary patency (%)Beathard 1999 39 AVL 82 -- 75Faiyaz 2002 17 AVL 88 -- --Planken 2007 10 AVL 89 -- --Mallik 2011 50 PNA -- 78 87Long 2011 21 PNA 90 71 95Bharat 2012 54 SLOT 83 -- --Lee 2013 31 PNA/AVL 87 -- 86Mufty 2015 31 PNA/AVL 94 68 85Nikam 2015 41 Optiflow 76 78 (3 mts) --Sadaghianloo 2016 53 RADAR 92 93 (6 mts) 100 (6 mts)Chemla 2016 20 Laminate 74 79 (6 mts) --

PNA= proximal neo-anastomosis; AVL=accessory vein ligation; SLOT= Straight Line Onlay Technique ; RADAR= Radial Artery Deviation And Reimplantation; Optiflow=internal anastomotic device; Laminate=external anastomotic device. Adapted from Tordoir et al. [53]

of 226

Chapter 4. Self-administered interventions for AV fistula maturationAuthorTitlePublication YearJournal

Study Design N CentresLocation


Patient Characteristics Comparator Intervention N Patients

SalimiAssessment of effects of upper extremity exercise with arm tourniquet on maturity of arteriovenous fistula in hemodialysis patients2013Journal of Vascular Access[60]


1Iran

Not reported Inclusion: Referred for brachiocephalic AV fistula construction on non-dominant arm; male&female; ability to exercise correctlyExclusion: age < 14 years; previous brachiocephalic or distal AV fistula; central venous stenosis; atherosclerotic vascular disease; arterial diameter < 2mm; Body Mass Index < 21 or > 25 kg/m2; unable to exercise correctly

Typesimple hand exercise with opening and closing the fingersDosenot reported

Typestructured isometric exercise programmeDose3 days a week, 4 times a day commenced 2 days after AV fistula surgeryDuration2 weeks

Enrolled: 55Actually Started: 55Included Comparator Group: 25Included Intervention Group: 30Analysed: 50

AgeMale Sex %







Agecontrol 51 years (±20 years); intervention 51 years (±13 years)Male Sex %control group 76%; intervention group: 84%

NoCommentAccording to the patients? file number, patients were randomly allocated into two groups' - patient's file number not explained

UnclearCommentnot stated

NoCommentblinding to exercise not possible


YesCommentA single radiologist who was not aware of patient group assessed all patients

NoCommentFive patients (17%) in the case group did not correctly follow the exercise programme and were excluded from the analysis.

The authors declare no conflict of interest or financial support.

Outcome Defined As Time Point of the Outcome





Effect Measure


Sonographic maturation

Draining vein ≥ 6mm, skin-vein distance ≤ 6mm, blood flow ≥ 600 ml/min measured 2 cm from anastomosis

2 weeks 17 22 25 25 not reported

0.14 Low, AV fistula maturity does not refer to successful dialysis, which is the commonly used measure of maturity.

Clinical maturation

Easily palpable, relatively straight with > 10 cm length, uniform thrill determined by nurse

2 weeks 5 13 25 25 not reported

0.008 Low, very subjective definition of maturity.

of 226


Study Design

N CentresLocation



KongThe effect of two different hand exercises on grip strength, forearm circumference, and vascular maturation in patients who underwent arteriovenous fistula surgery2014Annals of Rehabilitation Medicine[61]


1Korea

Not reported Inclusion: patients who received an AV fistula not further describedExclusion: injury, history of surgery, or limited mobility in the fistula-created limb, peripheral neuropathy, radiculopathy, rheumatic disorders, myopathy, arthritis

Typesoft ball for hand-squeezing exerciseDose60 squeezes twice dailyDuration4 weeks

Typehandgrip for hand-squeezing exerciseDose60 squeezes twice dailyDuration4 weeks


AgeMale Sex %









unclearCommentnot stated







Outcome Defined As






Effect Measure

P-Value

Evidence Certainty

No outcome relevant to the guideline reported

- - - - - - - - Very low, selective outcome reporting of surrogate endpoints, maturation not reported





FontseréEffect of a postoperative exercise program on arteriovenous fistula maturation: A randomised controlled trial2016Hemodialysis International[62]


1Spain

2013-2014 Inclusion: Outpatients with CKD 5&5D, candidates for AV fistula; ability to understand and perform the exercise programExclusion: previous dysfunctioning AV fistula in same arm, AV grafts, known arterial or central venous diseases in same arm, living far from hospital

Typeno exerciseusual lifestyle

Dose-Duration1 month

Typeexercise program: elbow and wrist flexion-extension, hand open-closeDose2 sets of 10-25 reps/dayDuration1 month


of 226

AgeMale Sex %









YesCommentautomatic randomization system based on Efron randomization (to ensure a similar distribution of patients in two groups in small trials, ratio 1:1)


NoComment the patients themselves knew the results of the randomisation

NoComment the nurse who is instructing the patients to follow the exercise knew the results of the randomisation

YesCommentspecialists dedicated to this study were blinded to the results of randomization in all phases of the study.

NoCommentoutcomes measured for 69/73 participants







Effect Measure and value

P-Value

Evidence Certainty

Clinical maturation

an easily palpable vein, with a straight-superficial segment, length > 10 cm, sufficient diameter, and good palpable thrill

1 month 25 36 31 38 Risk difference 0.14

0.07 no, risk of selection bias, wide confidence intervals and surrogate outcome

Ultrasonographic maturation

draining vein diameter ≥5 mm, skin–vein distance ≤6 mm, and brachial blood flow rate ≥500 mL/min

1 month 23 31 31 38 Risk difference 0.07

0.46 no, risk of selection bias, wide confidence intervals and surrogate outcome


Study Design

N CentresLocation



DesaiEarly application of an intermittent pneumatic compression device is safe and results in proximal arteriovenous fistula enlargement2018Journal of Vascular Access[63]


1India

2016-2017 Inclusion: Haemodialysis via central venous catheter; candidate for upper arm radio-cephalic or brachiocephalic AV fistula.Exclusion: Active infection, skin disorders or previous failed AV graft in ipsilateral arm;Restless arm or low systemic blood pressure; Poor pain tolerance; 4. Poor understanding of AV access, study protocol, poor family support.

TypeSham device

Dose6 hours/day

Duration4 weeks

TypeFist Assist - device to apply reliable intermittentpneumatic compression to the outflow veins 1 week after successfulAV fistula creationDose6 hours/dayDuration4 weeks


of 226

AgeMale Sex %







Agenot reportedMale Sex %control group 69%; intervention group: 77%

UnclearCommentmethod not reported

UnclearCommentnot reported

YesCommentUse of sham device

NoComment'single-blind randomised trial'

UnclearCommentnot reported

NoCommentoutcomes measured for 46/48 participants

Dr Tej M. Singh is the President, CEO of Fist Assist®Devices, LLC, which owns the device used in the study. Dr Singhowns the intellectual property associated with the Fist Assistdevice








Outflow vein dilation at 5 cm

% increase in vein diameter at 5 cm proximal to the arteriovenous anastomosis

1 month 25.1 45.2 16 30 Risk difference 20.1

0.026 Low, small study, wide confidence interval



1 month 42.1 49.3 16 30 Risk difference 7.2




1 month 41.5 38.8 16 30 Risk difference

-3.3


of 226

Chapter 5. Perioperative prophylactic antibiotics for preventing AV access infectionAuthorTitlePublication YearJournal





BennionA randomised, prospective study of perioperative antimicrobial prophylaxis for vascular access surgery1985Journal of Cardiovascular Surgery[64]


North America1

Not reported Inclusion: placement of expanded polytetrafluoroethylene (PTFE) grafts for vascular access; failed Cimino fistula or vessels unsuitable for an autogenous AV anastomosisExclusion: current infection; previously infected vascular access graft

Type: 0.9% salineDose: just before surgery, 6 and 12 hours after surgery

Type: cefamandoleDose: 1 g IVDuration: just before surgery, 6 and 12 hours after surgery

UnclearComment: Each patient was assigned randomly to one of two groups

UnclearComment: not reported




Outcome measured for all part.?

Percent lost to follow-up?

Reasons for lost to follow-up?


N Patients Age

Male Sex %YesComment: The control group received a placebo of 0.9% saline alone, under the same dosage schedule.



NoComment: 2/40 (5%) patients lost to follow-up

5% Requested transfer to other dialysis unit

No Evaluated for inclusion: not reportedEnrolled: 40Actually Started: 40Included Comparator Group: 20Included Group 1: 20


95%






P-Value

Evidence Certainty

Judgement Comments

Superficial cellulitis overlying the graft Not reported 4 1 19 19 Subdermal infection without graft involvement

Not reported 1 0 19 19

Infection of the prosthetic graft itself Not reported 3 1 19 19 Sum of all patients with class i, ii, iii infection

Not reported 8 2 19 19 0.04 Moderate, there are limitations

Old study, inadequate randomization sequence generation, radio-cephalic and femoral position of grafts mixed

of 226


Study Design

LocationN Centres



Comparator Type



ZibariPreoperative vancomycin prophylaxis decreases the incidence of hemodialysis vascular access infections. [abstract]1995Journal of the American Society of Nephrology[65]


North America1

Not reported

Not reported No antibiotics, but not further specified

Type: vancomycin IV Dose: 750 mgDuration: single dose 6 to 12 hours before vascular access placement procedure

NoComment: Randomization based on patient's hospitalization numbers.

NoComment: where even numbers were assigned to group 1 and odd numbers and odd numbers to group 2




Outcome measured for all participants?

Percent lost to follow-up?



N Patients Age

Male Sex %

Kidney Function

NoComment: Blinding must be assumed unsuccessful if group allocation determined by a simple rule of odd and even numbers

NoComment: The investigators were not blinded

NoComment: Outcome assessor generally not the same individual involved in surgery', but blinding must be assumed unsuccessful due to allocation method.before the surgical procedure

Not reported, unclear as the number of procedures rather than the number of patients was unit of analysis

Not reported No The 206 patients underwent 408 vascular access procedures

Evaluated for inclusion: not reportedEnrolled: 206Actually Started: 206Included Comparator Group: number of patients not reported, but 202 proceduresIncluded Group 1: number of patients not reported, but 206 procedures

54 years (17 years -81 years)53.5 years

43%

Hemodialysis





Anal. Participants Intervention Group 1

Effect Measure




P-Value

Evidence Certainty

Judgement Comments

14 to 30 days depending on timepoint of first use of graft

12 2 Number of patients not reported, but 202 procedures

Number of patients not reported, but 206 procedures

Mantel-Haenszel estimates of relative risk

0.16 0.05 0.59 0.006 Moderate, there are limitations

1. 46% of the procedures were secondary including thrombosis repair and revisions.2. Unit of analysis is number of procedures, not number of patients.3. All infections occurred in grafts, none in fistulas.

of 226


AV Fistulas





RaynerCreation, cannulation and survival of arteriovenous fistulae: data from the Dialysis Outcomes and Practice Patterns Study2003Kidney International[68]

Prospective cohort study

Global309

1996-2001 InclusionIncident patients starting haemodialysis via an AV fistula F (without prior temporary access) and first cannulation of an AV fistula < 730 days after creation; analysis at patient level ExclusionAge < 17 years

Intervention Comparator Exposed cohort representative?

Non-exposed cohort from same community?

Exposed ascertained via secure record?

Outcome not present at start?

Most important confounder adjusted?

Control for additional confounder?

Primary outcome adequately assessed?

Follow-up long enough for outcomes?

First cannulation < 14 days after creation

First cannulation > 14 days after creation

Yes Yes Yes No Yes Yes Yes Unclear

Follow-up complete for all subjects?

N Patients Age Outcome Defined As Analysed Participants Intervention Group


Effect Measure


Comments

Unclear Evaluated for Inclusion: 3674Enrolled: 3674Included: 894Included Comparator Group: 136Included Intervention Group: 758

Not reported

Fistula failureAny event resulting in non-function AV access

758 136 Relative Risk

2.10, p=0.006

No significant trend to decreased fistula failure among cannulation intervals >14 days

of 226


Study Design





Non-exposed cohort from same comm?

Expositionascertained via secure record?


SaranTiming of first cannulation and vascular access failure in haemodialysis: an analysis of practice patterns at dialysis facilities in the DOPPS2004Nephrology Dialysis Transplantation[69]


Global309

1996-2001 Inclusionnew AV access: 2730 grafts2154 AV fistula

centre level: # weeks after AV access insertion

AV grafts<2 weeks3-4 weeks>4 weeks

AV fistulas<1 month2-3 months>3 months

centre level: # weeks after AV graft insertion

AV grafts2-3 weeks

AV fistulas1-2 months

Yes Unclear Yes Yes





N Patients Age Analysed Participants Intervention

Follow up

Yes Yes Yes Unclear 48842730 grafts2154 AV fistulas

60 years 27302154

Mean 4 months

Outcome Defined As


Effect measure Outcome Certainty evidence

Access failureTime to first thrombosis or access salvage procedure

- Hazard Ratio AV grafts<2 weeks: 0.84, p=0.113-4 weeks: 0.94, p=0.48>4 weeks: 0.93, p=0.48

AV fistulas<1 month: 0.72, p=0.082-3 months: 0.91, p=0.43>3 months: 0.87, p=0.31

Very Low, centre level data



Patient Characteristics Intervention Comparator Exposed cohort representative?


MedkouriAnalysis of Vascular Access in Hemodialysis Patients: A report from a dialysis unit in Casablanca2006Saudi Journal of Kidney Diseases and Transplantation[70]

Retrospective cohort study

Africa1

InclusionPatients who need dialysis treatment

First cannulation ≤ 21 days after creation


Yes Unclear

Most important confounder

Primary outcome

Follow-up complete for all

Additional Sources of Bias N Patients Age Outcome Defined As


of 226

adjusted? adequately assessed?

subjects?

No Unclear Unclear Very young patients. Data from Africa (Casablanca)

Evaluated for Inclusion: 190Enrolled: 190


ThrombosisNot otherwise defined

Authors state no difference between < and > 21 days before cannulation, however no data provided.






CulpVascular access thrombosis in new hemodialysis patients1995American Journal of Kidney Diseases[71]


North America26

InclusionPeople newly diagnosed with end-stage renal diseaseExclusionocclusion of AV fistula or graft before first cannulation

First cannulation of graft ≥ 30 days after AV access creation

First cannulation < 30 days after AV access creation

Unclear No Yes



N Patients Age OutcomeDefined As

Effect measure Value effect measure

Yes Unclear Evaluated for Inclusion: not reportedEnrolled: 267AV grafts: 147AV fistulas: 118

62 years ± Not reported

Thrombosisthrombosis or intervention within 1 year after first cannulation

Relative risk PTFE grafts: 0.77; 95%CI 0.43 to 1.38

AV fistula: 0.40; 95%CI 0.19 to 0.84


Study Design



Patient Characteristics Intervention Comparator

RavaniCardiovascular Comorbidity and Late Referral Impact Arteriovenous Fistula Survival: A Prospective Multicentre Study2004Journal of the American Society of Nephrology[72]


Europe3

1997-2002 Inclusionall consecutive ESKD patients>18 years,new AV fistulamaintenance hemodialysis treatment programs ExclusionRepeated vascular accessAV fistulas created by others than the local renal physicians in charge of the vascular access-related procedures

First cannulation < 30 days after creation


of 226

Exposed cohort representative?







N Patients Age Outcome Defined As

Effect Measure


Yes Yes Yes Yes Yes Yes Yes Evaluated for Inclusion: 535Enrolled: 513Actually started:446

67 years

Intervention-free period to first failure

Hazard Ratio

1.94; 95%CI 1.34 to 2.82



Start/End

of Study (year)





WilminkIs early cannulation of an arteriovenous fistula associated with early failure of the fistula?2017Journal of Vascular Access[73]


United Kingdom1 Hospital Trust with 7 dialysis programs

2002 - 2016

Inclusionnot reportedExclusionnot reported

Cannulation < 30 days

Cannulation ≥ 30 days

Yes Yes Yes







Financial support?

additional sources of bias?

N Patients Age

Yes No No Yes Unclear 23% lost to follow-up

No 1. Unclear what factors influenced the decision or who made the decision upon timing to cannulate2. Unit of analysis is fistulas rather than patients3. Unadjusted comparison of early and late cannulation

Not reported n refers to fistulaeEvaluated for inclusion: 1641Included: 1167Included comparator group: 1019Included intervention group: 148

Not reported

Outcome Defined As time point of outcome assessment



Effect Measure Value Effect Measure

Evidence Certainty Comments

Early failuredefined as abandonment within 90 days of first cannulation

Not reported Not reported Not reported Chi square P=0.35 Very low 1. Unadjusted analysis2. Unit of analysis = fistulas3. Same cohort as in Wilmink 2017 NDT

of 226








WilminkEffect of first cannulation time and dialysis machine blood flows on survival of arteriovenous fistulas2017Nephrology Dialysis Transplantation [74]


United Kingdom1 Hospital Trust with 7 dialysis programs

2002 - 2016 Inclusionnot reportedExclusionnot reported

Cannulation within 4 weeks after AV fistula creation

Cannulation > 4 weeks

Yes Yes Yes







Financial support?

Additional sources of bias? N Patients Age

Yes No No Yes Unclear 23% lost to follow-up

Not reported 1. Unclear what factors influenced the decision or who made the decision upon timing to cannulate2. Unit of analysis is fistulas rather than patients3. Unadjusted comparison of early and late cannulation

Not reported n refers to fistulaeEvaluated for inclusion: 1641Included: 1167Included comparator group: 1053Included intervention group: 114

Not reported

Outcome Defined As Time point of outcome assessment



Effect Measure


Evidence Certainty

Comments

Fistula survivaldefined as cumulative patency from AV fistula creation until abandonment irrespective of any interventions

Not reported Pre-dialysis patients: 34maintenance dialysis patients: 80

Pre-dialysis patients: 502maintenance dialysis patients: 551

Log rank test Pre-dialysis patients: p= 0.88maintenance dialysis patients: p=0.19

Very low 1. Unadjusted analysis2. Unit of analysis = fistulas3. Same cohort as in Wilmink 2017 J Vasc Accesssmall sample sizes in subgroups


Study Design


Start/End

of Study (year)





Allon2018Relationships between clinical processes and arteriovenous fistula cannulation and maturation: a multicenter prospective cohort studyAmerican Journal of Kidney Diseases


USA7

2010-2013

Inclusionadults with ESKD and treated with haemodialysis set to undergo AV fistula creation - single stage upper extremityExclusion

Cannulation at 8 weeks after creation

Cannulation > 8 weeks

Yes Yes Unclear

of 226

[75] not reported







Financial support?

Additional sources of bias?

N Patients Age

Yes No Yes Yes Yes Unclear Not reported

No standardised decision rules to guide first cannulation attempt

Not reported n refers to fistulaeEvaluated for inclusion: 602Included: 387Included comparator group: not reportedIncluded intervention group: not reported

Not reported

Outcome Defined As

Time point of outcome assessment



Effect Measure


Evidence Certainty

Comments

Overall maturationdialysis with 2 needles for≥75% of haemodialysis sessions during a continuous 4-week period commencing within 9 months of AVF surgery and including 4 consecutive sessions with mean blood pump flow >300 mL/min or, failing that, any Kt/V≥1.4 or urea reduction ratio ≥70%

Time-to-event Not applicable Not applicable Odds Ratio (per additional month)

0.93 - 95%CI 0.89 to 0.98

Low Specific analysis one of many primarily aimed at assessing clinical processes, timing of first cannulation and successful dialysis. No adjustment for multiple testing. No adjustment for clinical appreciation of readiness of AV fistula for haemodialysis; likely insufficient adjustment for other variable that may influence choice to postpone cannulation and later successful dialysis.


Study Design



Patient Characteristics Comparator

FerringAccuracy of early postoperative clinical and ultrasound examination of arteriovenous fistulae to predict dialysis use2014Journal Vascular Access[76]


Europe1

2008-2009 Inclusionpatient undergoing 1st or 2nd vascular access creationExclusion1/access occlusion within the first 4 weeks; 2/patient undergoing salvage or >2nd attempt

clinical examination included the flow character (categorised as thrill, pulse, audible bruit without a palpable thrill), the vein calibre, the straightness and depth of the vein

Intervention 1 Intervention 2 Exposed cohort representative?




Ultrasound: Representative cross-sectional vascular lumen diameters were obtained at the following sites: brachial, radial and ulnar arteries, AV fistula anastomosis, curved vein beyond anastomosis, proximal straight vein and likely

Ultrasound: Velocity measurements (peak systolic and end-diastolic velocities) were obtained at the following sites: brachial, radial and ulnar arteries, AV fistula anastomosis, curved vein beyond anastomosis, proximal

Yes Yes Yes No

of 226

future cannulation site straight vein and likely future cannulation site

Most important confounder adjusted? Control for additional confounder?





No. Comment: no adjustment for forearm versus upper arm site of AV fistula (p<0.11, Table 1)

No Yes Yes No. Comment: Of the 151 patients with patent AV fistula 119 had known outcomes

21% (32/151). Reason: no clinical or ultrasound examination or AV fistula outcome unknown at the end of study (not started dialysis)

Financial interest for sponsor avoided?

N Patients Age Outcome Defined As Outcome Value Comparator Group



No Evaluated for Inclusion: 151Enrolled: 133Included Comparator Group: 119Included Group 1: 106Included Group 2: 116Included Group 3: 116


AV fistula dialysis use, which was defined as AV fistula that could be used for at least six consecutive haemodialysis sessions, with two needle AV fistula cannulation and a blood flow rate >200 mL/min

Thrill -pulse-audible bruit without a palpable thrill: Sensitivity 96% Specificity 21% Positive predictive value 81% Negative predictive value 63%

Vein diameter in the straight part of the outflow vein >5 mm: Sensitivity 83% Specificity 68% Positive predictive value 90% Negative predictive value 53%

Arterial velocity (end-diastolic velocity) in the proximal artery <110 cm/sec: Sensitivity 67% Specificity 65% Positive predictive value 87% Negative predictive value 37%










JaberiArteriovenous fistulas for hemodialysis: application of high-frequency US to assess vein wall morphology for cannulation readiness2011Radiology[77]


North America1

2008-2010 Inclusion: patients needing a newly created native AV fistula

Unclear Yes Yes No Unclear





Fin interest for Sponsor avoided?

N Patients Age Outcome Defined As



Unclear Yes Yes Yes No Evaluated for Inclusion: 23Enrolled: 20Included Comparator Group: 20Included Group 1: 20Included Group 2: 20

59 years No extravasation at either the proximal (arterial) or distal (venous) puncture site during the first dialysis treatment

50% of patients and 62% of cannulation procedures had no extravasations

Ultrasound 1 week prior to cannulation with linear 40-55 MHz probe: anterior venous wall intima-media-thickness >0.13 mm was significantly (p<0.001) associated with successful cannulation with a sensitivity of 86.7% and a

of 226

Included Group 3: 20 specificity of 91.7 %.





RobbinHemodialysis arteriovenous fistula maturity: US evaluation2002Radiology[78]


North America1

1998-2000 Inclusion Ultrasound performed within 4 months after fistula placement

Exclusion Ultrasound performed for any indication other than assessment of fistula maturation, such as fistula mass or hematoma evaluation

1/venous diameter >0.4 cm by ultrasound

2/blood volume flow >500 mL/min by ultrasound

Clinical evaluation by registered dialysis nurses with > 5 years of experience in dialysis access techniques: 1/easily palpable superficial vein of adequate diameter 2/ a uniform thrill to palpation3/accessible draining vein needed to be relatively straight and more than 10 cm long









Financial interest for Sponsor avoided?

Yes Yes Yes No No Yes Yes NoComment: 15 patients had not started dialysis

No

N Patients Age Outcome Defined As Outcome Value Comparator Group




Evaluated for Inclusion: 122Enrolled: 95Included Overall:69Included Comparator Group: 54


Successful AV fistulaAV fistula supporting a blood flow rate of 350 mL/min for at least 6 dialysis sessions in 1 month

Clinical evaluationPositive predictive value: 17/21 fistulas (81%)Negative predictive value: 7/9 (78%) Overall accuracy of prediction was 80% (24 of 30)

Venous diameter >0.4cmPositive predictive value: 24/27 (89%)Negative predictive value: 12/27 (44%)

Blood flow >500 mL/minPositive predictive value: 26/31 (84%)Negative predictive value: 9/21 (43%)

Venous diameter >0.4cm + blood flow > 500ml/minPositive predictive value: 19/20 (95%)Negative predictive value: 5/15 (33%)




Patient Characteristics Comparator Intervention

WongFactors associated with early failure of arteriovenous fistulae for haemodialysis access1996European Journal of Vascular and Endovascular Surgery[79]


Europe1

1995-1996 InclusionAll patients with end-stage renal failure requiring a vascular access were invited to take part in this study. All created at the wrist with the anastomosis constructed in end vein-to-side artery fashion

Vein diameter at 2 weeks <4mm or fistula flow at 5-10cm from anastomosis <400 ml/min

Vein diameter at 2 weeks >4mm and fistula flow at 5-10cm from anastomosis >400 ml/min

of 226


Non-exposed cohort from same comm?

Exp ascertained via secure record?








Unclear Unclear Yes No Unclear Unclear Yes Yes Unclear 5% (3/60)

Additional Sources of Bias N Patients Age Outcome Defined As Time Point of the Outcome


Outcome Value comparator Group



Selected patient group as all wrist fistulas; not truly about first cannulation, so conclusion only valid in right direction: do not attempt cannulation if either vein diameter <4mm or flow <400ml

IncludedIncluded Comparator Group: 14Included Group 1: 40

59 years Successful AV fistulas

fistulas used for routine dialysis

12 weeks 38 0 38 8

AV grafts




Patient Characteristics Intervention Comparator Sequence adequately generated?

SottiuraiComparative results of early and delayed cannulation of arteriovenous graft in haemodialysis1997European Journal of Vascular and Endovascular Surgery[80]


North America1

Not reported Inclusionpatients with AV graft insertion requiring urgent haemodialysis

Exclusionnot reported

Early cannulation 1-2 days after AV graft insertion

Late cannulation 10-14 days after AV graft insertion

Unclear

Commentrandomisation procedure not reported





Outcome measured for all participants?



N Patients Age

Male Sex %


Unclear

Commentrandomisation procedure not reported

No

Commentnot reported, impossible due to nature intervention

No

Commentnot reported, probably not due to nature intervention

Unclear

Commentnot reported

Yes 0

Commentno information provided

Yes Enrolled: 36Intervention: 18Comparator: 18

51 years +/- Not reported

47%

Diabetes: 19%

OutcomeDefined As



Outcome Value Comparator group



Effect Measure

Value Certainty results

Perigraft haematomaNot otherwise defined

40 days 1 3 18 18 Relative Risk 0.39, 95%CI 0.11 to 1.37

Moderate, small study, wide confidence interval

of 226

Cellutitis 40 days 2 1 18 18 Relative Risk Not estimable Low, small study, few eventsThrombosis 40 days 1 1 18 18 Relative Risk 1.00, 95%CI 0.45 to

2.23Moderate, small study, wide confidence interval

Pseudoaneurysm 40 days 0 0 18 18 Relative Risk Not estimable Low, small study, few eventsCompression time to control haemorrhage (min)

First cannulation

6 ± 1.9 6.6 ± 4.2 18 18 Mean Difference

-0.50, 95%CI -2.63 to 1.63

Moderate

Venous back pressure (mmHg)

First cannulation

141 ± 36 135 ± 41 18 18 Mean Difference

6.00, 95%CI -19.21 to 31.21

Moderate





FeldmanEffect of timing of the first cannulation on survival of arteriovenous hemodialysis grafts2013Therapeutic Apheresis and Dialysis[81]


Israel1

2007-2010 Inclusion-Age 18-85 years-ESKD, treated with chronic hemodialysis-Insertion of PTFE AV graft, successful first cannulation

ExclusionThrombophilia; AV grafts with major complication of graft surgery; AV grafts that underwent any intervention before the successful first cannulation and sophisticated extra-anatomic AV grafts: e.g. femoral artery to inferior vena cava bypass

Cannulation week 2 after AV graft insertion

Cannulation week 3->7 after AV graft insertion










N Patients Age

Unclear Yes No No No Yes Yes No 1/ All patients with early cannulation had reasons for this e.g. catheter infection: bias by indication; 2/ only "uneventful grafts" were included

Evaluated for Inclusion: 87Enrolled: 87Analysed: 58Intervention: 51Comparator group: 7


of 226

Outcome Defined As


Outcome Value Intervention

Outcome Value Comparator

Effect measure

Value Certainty evidence

Successful cannulation 100% 100% - Not estimable Very low, bias by indicationPrimary AV graft failurethe first occurrence of graft thrombosis or any access procedure performed to maintain or re-establish patency of the graft

12 months 7/7 (100%) Week 3: 7/9 (78%)Week 4: 6/7 (86%)Week 5: 8/12 (67%)Week 6: 6/9 (67%)Week 7: 5/10 (50%)Overall

Hazard Ratio

1.63, CI 0.65 to 5.920.93, CI 0.28 to 3.101.49, CI 0.47 to 4.721.17, CI 0.39 to 3.520.89, CI 0.27 to 2.931.53, CI 0.15 to 1.85All Hazard Ratio compared with overall hazard

Very low

Cumulative AV graft failure 12 months 4/7 (57%) Week 3: 2/9 (22%)Week 4: 2/7 (29%)Week 5: 6/12 (50%)Week 6: 4/9 (45%)Week 7: 4/10 (40%)

Hazard Ratio

1.97, CI 1.46 to 8.370.36, CI 0.32 to 3.931.49, CI 0.25 to 8.971.24, CI 0.23 to 6.760.70, CI 0.10 to 5.001.60, 0.83 to 4.24

Very low







DawidsonEarly Use of the Gore-Tex Stretch Graft1996Blood Purification[82]


North America1

1991-1993 InclusionFirst graft: 4-7 mm tapered vascular graft

ExclusionNot reported

First cannulation <14 days after AV graft insertion

First cannulation >14 days after AV graft insertion

Unclear Unclear No





Additional Sources of Bias N Patients Age Outcome Defined As



No Yes Yes Unclear - Early cannulation especially in patients without other access and urgent start of haemodialysis- Large proportion of minority patients in both groups (85% respectively 81%)

Evaluated for Inclusion: 270 Included Comparator Group: 174Included Intervention group 1: 96

50 years AV graft survivalSecondary patency

1 year: 80%2 years 78%3 years 73%

1 year: 93% 2 years 92%3 years 91%(p=0.008)

Failures caused by infection

7% 1% (p<0.01)

of 226


Study Design



Patient Characteristics Intervention Comparator Exposed cohort represen-tative?



GlickmanProspective multicenter study with a 1-year analysis of a new vascular graft used for early cannulation in patients undergoing hemodialysis 2015Journal of Vascular Surgery[83]



2010 -2013 Inclusioncurrently undergoing hemodialysis or expected to commence within 30 days; not considered for fistula creation; suitable for graft creation in upper extremity determined by surgeon and based on vein mappingExclusion>2 previous vascular accesses in the same arm; systemic infection; bleeding disorder; hypercoagulopathy; on maintenance immunosuppression; treatment with extended-release dipyridamole plus aspirin

Early cannulation within 72 hours after implantation of a multilayer ePTFE graft (Gore ACUSEAL Vascular Graft; W. L. Gore& Associates, Flagstaff, Ariz)

Cannulation > 21 days after placement

Yes Yes Unclear







Financial support? Additional sources of bias?

N Patients Age

Yes No No Unclear Yes Unclear 1. This study was sponsored by W. L. Gore & Associates.2. All authors have served as consultants to W. L.Gore&Associates.

1. Timing of first cannulation decided by responsible physician.2. Unadjusted analysis

Evaluated for Inclusion: not reportedIncluded: 138Included Comparator Group: 33Included Intervention Group: 54


Outcome Defined As Time point of outcome assessment



Effect Measure

Value Effect Measure Evidence Certainty

Comments

Cumulative patency ratedefined as freedom from complete loss of the access regardless of whether interventionswere done to restore or to maintain patencyor to manage hematoma, infection, or steal syndrome during the 12-month study period

Up to 12 months

54 33 1. Cumulative rates2. Log rank test

1. Early cannulation: 76% (95% CI 62% - 86%) vs late cannulation: 77.5% (95% CI 58%-89%)2. p= 0.7

Very low 1. Small sample size, further reduced by arbitrary cut-points to compare timing of first cannulation 2. Selected timeframes leave a gap for first cannulation occurring between 73 hours and <21 days following graft placement3. Timing of first cannulation decided by responsible physician not an objective rule4. Unadjusted analysis5. Study supported by industry

Primary unassisted patencydefined as freedom from complete loss of the accesswithout thrombosis or an intervention during the 12-month period

Up to 12 months

54 33 1. Cumulative rates2. Log rank test

1. Early cannulation: 30% (95% CI 18% - 43%) vs late cannulation: 41% (95% CI 24%-57%)2. p= 0.7

of 226


Study Design

LocationN centres




HakaimDurability of early prosthetic dialysis graft cannulation: results of a prospective, nonrandomised clinical trial 1997Journal of Vascular Surgery[84]


North America1

1992-1996 InclusionPatients receiving brachial artery to axillary vein stretch-expanded-PTFE graft not further specifiedExclusionnot reported but no patient had a history of ipsilateral central venous catheterization, warfarin therapy, or a previous AV fistula at the proposed operative location

Early cannulation with a 16-gauge needle, 24 to 72 hours after AV graft insertion if dialysis initiation was needed

Late cannulation with a 15- or 16-gauge needle 14 days after surgery

Unclear because early cannulation depended on requirement for early dialysis rather than access criteria

Yes









Yes Yes No univariate analyses only

No univariate analyses only

Yes Yes Yes Patients in early cannulation group had a reason for urgent start dialysis, so are likely different from those with late cannulation

Evaluated for Inclusion: not reportedEnrolled: 76Included late cannulation: 30Included early cannulation: 46





P-value Comment

Graft thrombosis Before cannulation 1/30 (3.3%) 1/46 (2.2%) Not significant Primary patency not otherwise defined

3 months 97% 94% Not significant Outcome values reported for number of grafts created rather than number of patients

Primary patency not otherwise defined

12 months 74% 70% Not significant Outcome values reported for number of grafts created rather than number of patients

Cumulative patient survival not further defined Not reported 74% 67% 0.50

of 226



Study characteristics Comparator Intervention Amstar Score

N studies included

N Patients included

Age

RavaniPre-emptive correction for haemodialysis arteriovenous access stenosis2016Cochrane Systematic Review[87]

InclusionRCTs; quasi-RCTs of pre-emptive stenosis correction in a mature AV access (fistula and graft) evaluating the effects of strategies planned to identify (through regular access surveillance) and correct pre-emptively a previously unknown/unidentified access stenosis (primary prophylaxis) or the effects of pre-emptive correction of a known stenosis in a non-dysfunctional access (secondary prophylaxis).Exclusionhemodialysis for acute kidney injury, central venous catheters, studies comparing different approaches to salvage a dysfunctional, failing or clotted access

Deferred correction of stenosis

Pre-emptive correction

11/11 14 1393 59 years average across studies in review

OutcomeDefined As








Evidence Certainty

DeathAll-cause mortality 30 43 192 194 Relative Risk 1.38 0.91 2.11 Very low

Access lossAccess loss overall (for grafts and fistulas combined)- all studies 122 106 461 511 Relative Risk 0.81 0.65 1.02 Low

Access loss for fistulas only 34 17 154 156 Relative Risk 0.50 0.29 0.86 LowAccess loss for grafts only 88 89 307 355 Relative Risk 0.90 0.71 1.15 LowAccess loss for primary prophylaxis 84 79 333 376 Relative Risk 0.84 0.65 1.10 LowAccess loss for secondary prophylaxis 38 27 128 135 Relative Risk 0.75 0.49 1.13 LowFistula loss by primary prophylaxis 10 4 58 53 Relative Risk 0.44 0.15 1.31 LowFistula loss by Secondary prophylaxis Studies 24 13 96 103 Relative Risk 0.52 0.28 0.97 LowAccess loss by surveillance method using Doppler ultrasound (access flow and structure) 71 61 263 285 Relative Risk 0.81 0.61 1.09 Low

Access loss for surveillance method using blood flow (access flow) only 13 18 70 91 Relative Risk 0.98 0.52 1.86 Low

Access thrombosisAccess thrombosis overall 199 187 566 646 Relative Risk 0.79 0.65 0.97 LowAccess thrombosis Grafts only 130 150 318 379 Relative Risk 0.95 0.80 1.12 ModerateAccess thrombosis Native fistulas only 69 37 248 267 Relative Risk 0.50 0.35 0.71 Moderate

of 226

Access thrombosis by Primary prevention 125 145 406 479 Relative Risk 0.94 0.78 1.12 LowAccess thrombosis by Secondary prophylaxis 74 42 160 167 Relative Risk 0.53 0.32 0.84 LowFistula thrombosis by Primary prophylaxis 22 17 152 164 Relative Risk 0.75 0.42 1.36 LowFistula thrombosis by secondary prophylaxis 47 20 96 103 Relative Risk 0.40 0.26 0.62 Low Access thrombosis by surveillance method Blood flow (access flow only) 55 78 214 256 Relative Risk 1.08 0.83 1.40 Low

Access thrombosis by surveillance method using Doppler ultrasound (access flow & structure)

65 63 171 188 Relative Risk 0.84 0.65 1.08 Low

Access thrombosis by surveillance method using static venous pressure only 5 4 21 35 Relative Risk 0.54 0.17 1.70 Low

InfectionAny infection as defined in the original study 123 125 Incidence Rate Ratio 1.74 0.78 3.91 Very low

ProceduresAngiograms 243 296 Incidence Rate Ratio 1.78 1.18 2.11 Moderate

CatheterCatheter use 184 210 Incidence Rate Ratio 0.58 0.35 0.98 Low

HospitalisationHospitalisation 94 125 Incidence Rate Ratio 0.54 0.31 0.93 Low





AragoncilloAdding access blood flow surveillance reduces thrombosis and improves arteriovenous fistula patency: a randomised controlled trial2017Journal of Vascular Access[88]


5Spain

2012-2015 Inclusion: age >18 years, hemodialysis with functioning AV fistula ≥3 monthsExclusion: coagulopathy; hemoglobinopathy, hospitalization <3 months prior, or VA-related complications or dysfunction <3 months prior

Typeclassic surveillance:-predialysis physical examination-effective blood flow, dynamic arterial & venous pressure every dialysis-Kt/V 1x/week-recirculation/ 3 monthsDurationmedian 27 (range 11-35) months

Typeclassic surveillance + doppler ultrasound and doppler dilution methods/ 3 months

Durationmedian 25 (range 15-35) months

Enrolled: not reportedActually Started: 212Included Comparator Group: 105Included Intervention Group: 107Analysed: not reported

of 226

AgeMale Sex %

Sequence adequately generated? Allocation adequately concealed?






Agecontrol 66 years ±15 years; intervention 63 years ±15 yearsMale Sex %control group 68%; intervention group: 75%

YesCommentrandomisation using a computerised randomisation system based on 10-patient blocks


NoCommentopen-lable

NoCommentopen-lable

UnclearCommentopen-lable

NoCommentstaggered entry and censored for transplantation, death and transfer

The authors declare no conflict of interest. Financial support provided by the Infant Hospital Research Foundation









Thrombosis-free patency Median +/-26 months

Not reported Not reported 104 103 Hazard Ratio 0.38; 95%CI 0.11 to 0.82

0.01 Yes

Primary unassisted patency

Intervention-free access survival

Median +/-26 months


0.94 Yes

Secondary patency

Access survival until AV fistula abandonment

Median +/-26 months


0.03 Yes

Interventions Number of interventions during follow-up

Median +/-26 months

0.14/patient/year

'0.14/patient/year 104 103 Not reported 0.68 Yes

of 226

Chapter 8. Medical treatments for maintaining long term AV access patencyAuthorTitlePublication YearJournal

Study Design


Comparator

Intervention Amstar Score N studies included

N patients included

Age Outcome Outcome Defined As

TannerMedical adjuvant treatment to increase patency of arteriovenous fistulae and grafts2015Cochrane Database of Systematic Reviews[13]

Systematic review

InclusionRCTs of ESKD patients undergoing haemodialysis via an AV fistula or AV graftExclusionperitoneal dialysis patients, non RCTs

No adjuvant therapy

Adjuvant therapy

8/11 - duplicate study selection and data extraction not entirely independent; status of publication possibly inclusion criterion, and misinterpretation of lack of evidence for evidence of no effect

15

14 included in table, only long-term outcomes

2230 18 years -80 years

Graft/ fistula patency

Thrombosis


comparison Outcome Value Comparator Group




Effect Measure




Evidence Certainty

5 to 18 months after AV graft insertion

Aspirin vs placebo


0.62 0.05 7.16 Very low: 2 studies from before 1995, at unclear risk of bias, serious inconsistency I²=86%; indirectness (intermediate for access loss) and very serious imprecision

Up to 18 months after graft insertion

Dipyridamole vs placebo




Dipyridamole + aspirin vs placebo




Warfarin vs placebo


1.76 0.75 3.99 Low: 1 study, point estimate favouring placebo, but very imprecise.

12 months after AV graft insertion

Fishoil vs placebo

54 35 109 111 Odds Ratio

0.24 0.03 1.95 Very low: 2 studies at high risk of bias, high degree of inconsistency and very serious imprecision

6 months after AV fistula formation

Clopidogrel vs placebo


0.17 0.03 0.82 Low: 1 study at risk of bias, and serious imprecision with the confidence interval wide

6 to 12 months PRT-20l vs placebo

28 48 100 206 Odds Ratio

0.75 0.42 1.32 Very low: 3 studies, some at risk of bias and with very serious imprecision and confidence interval spanning line of no effect.

of 226


Study Design

Location Patient Characteristics Inclusion Comparator - TypeIntervention - Type

AMSTAR score



Comprehensive literature search?


PalmerAntiplatelet agents for chronic kidney disease2013Database of Systematic Reviews[14]

Systematic review

Global RCTS of CKD patients, including those who needed renal replacement therapy (dialysis)

Comparatorplacebo or no treatmentInterventionantiplatelet therapy

10/11 Yes Yes Yes Yes










N studies enrolled?


Yes Yes Yes Yes Yes yes, but not for vascular outcome

No 1093 182 50 27139


Study Design

Patient Characteristics Inclusion Comparator - Type

Intervention - Type

AMSTAR score



Comprehen-sive literature search?



ViecelliOmega-3 Polyunsaturated Fatty Acid Supplementation to Prevent Arteriovenous Fistula and Graft Failure: A Systematic Review and Meta-analysis of Randomised Controlled Trials2018American Journal of Kidney Diseases[15]

Systematic review

RCTs in which omega-3 polyunsaturated fatty acids were compared to placebo or no treatment to evaluate vascular access function in for prevention of vascular access failure in adult receiving or planning to receive haemodialysis using an AV fistula, AV graft, or arteriovenous shunt in upper or lower limb


Interventionomega-3 polyunsaturated fatty acids

10/11 Yes Yes Yes Yes No

Commentno list of excluded studies









N studies enrolled?



of 226







Effect Measure




P-Value

Certainty evidence

All-cause mortality

- 12 to 52 weeks



Very low4 RCTs Imprecision: risk estimate includes null effect Directness: data derived from small no. of studies in specific settings that may not be generalizable (treatment duration variable, access different; interventions may differ)

Primary patency failure

First thrombosis or need for surgical or endovascular intervention to restore patency

12 to 52 weeks


0.81 0.66 0.98 <0.05 Moderate3 RCTsDirectness: data derived from small no. of studies in different settings that may not be generalizable (treatment duration variable, access type variable)



12 to 52 weeks




Vascular access abandonment

Not reported 12 to 52 weeks



Low2 RCTsImprecision: risk estimate includes null effect Directness: data derived from small no. of studies in specific settings that may not be generalizable (treatment duration variable, access different;

of 226

interventions may differ)Bleeding Not

otherwise defined

12 to 52 weeks



Very low4 RCTs - not info about severity of bleeding Severe imprecision (2 downgrades): risk estimates include null effect and estimate consistent with both appreciable benefit and harm; Directness: data derived from small no. of studies in specific settings that may not be generalizable (treatment duration variable, dose variable, coadministration of antiplatelet agents variable)




AMSTAR score






WanEffects of far infrared therapy on arteriovenous fistulas in hemodialysis patients: a meta-analysis2017Renal Failure[17]

Systematic review

RCTs comparing Far-infra red therapy versus no treatment for promoting maturation in adults receiving haemodialysis using an AV fistula


InterventionFar infrared therapy

8/11 No Yes UnclearComment-many databases including several Chinese-Complete search strategy not provided

Yes NoCommentno list of excluded studies









N studies enrolled?









Effect Measure




P-Value

Certainty evidence

Primary patency


12 months



Low4 RCTs Risk of bias: unclear risk of biasData derived from small no. of studiesDirectness: only 5/21 contributed to meta-analysis. Quality of Chinese language trials difficult to assess.

of 226

AV fistula occlusion


12 months



Low5 RCTsRisk of bias: most studies at unclear risk of biasDirectness: only 5/21 contributed to meta-analysis. Quality of Chinese language trials difficult to assess. Definitions unclear


Study Design Location Patient Characteristics Comparator

Intervention 1

AMSTAR A priori design provided?




BasharRole of far infra-red therapy in dialysis arterio-venous fistula maturation and survival: systematic review and meta-analysis2014PLOS One[16]

Systematic review

Asia InclusionRCTs and observational studies to examine far-infrared therapy in ESKD patients with AV-fistula. Exclusioncase series and case reports.

Comparatorno treatment

Interventionfar infrared therapy

9/11 No Yes Yes Yes



Scientific quality studies



Conflict of Interest stated?

Evaluated for inclusion

N studies enrolled?


No

Commentonly list of included studies

Yes Assessed?YesUsed?Yes

Yes Yes Yes N citations1244N full papers8

4 Enrolled? 666Included Comparator Group? 326Included Group 1? 340

Age62 years +/-15 yearsMale Sex %52%

Outcome Defined As






Effect Measure




Results are reliable?

Primary patency: need for any interventional procedure (surgery or angioplasty) to correct an occlusive or malfunctioning fistula which could not sustain an extracorporeal blood flow < 200 mL/min during hemodialysis


1.23 1.12 1.35

Yes, but all studies from same centreSecondary patency: following salvage procedure 12 months 140 160 163 168 Relative

Risk1.11 1.04 1.19

of 226







AbacilarOral prostacycline analog and clopidogrel combination provides early maturation and long-term survival after arteriovenous creation: a randomised controlled study2015Indian Journal of Nephrology[23]


Turkey1

2008 - 2013

Inclusionpeople set to undergo AV fistula creation Exclusionactive or past bleeding; platelet count <75000/µL; coagulopathy; acute ulcer disease; blood pressure >200/115 mmHg; advanced liver disease; history of esophagitis or gastritis; discontinued antiplatelet; pregnancy or lactation; history of myocardial infarction; cerebrovascular accident in previous 12 months

Placebo Type + doseoral clopidogrel 75 mg once daily + oral prostacyclin analogue 200 mg once daily

Duration7-10 days before to one year after surgery

Yes

Commentrandomization stratified according to the medical centre with a permuted block scheme with block size = 4

Yes

Commentsequence was not known to the participant or personnel







Yes

Commentmatching placebo; medication not known to the participant

Yes

Commentmedication not know to personnel

Yes

Commentparticipants of the study team were blinded to the treatment

Yes 0% Yes Evaluated for Inclusion?385Enrolled? 96Included Comparator Group? 46Included Group 1? 50

Age55 years +/-1 year

Male Sex %68%

OutcomeDefined As






Effect Measure

Value 95%CI, P-value

Certainty of the evidence

Maturationblood flow ≥ 300 ml/min, velocity ≥ 70 cm/sec


1.28 1.01 to 1.61 Low. Proportion maturation such defined decreased over time; reported as proportions, denominators are unclear; low external validity due to exclusion criteria

Primary AV fistula failurenew fistula creation


0.26 0.09 to 0.74 High, but low external validity due to exclusion criteria

Bleeding 1 year Not reported 0 46 50 Not reported

Not reported

Not reported Low, all patients prone to bleeding were excluded from the study

of 226


Study Design

LocationN Centres



Comparator - Type



BleyerThe effect of topically applied Vonapanitase on fistula outcomes: results of the PATENCY-1 Trial2017American Journal of Kidney Diseases(Abstract)[24]



Not reported

Inclusionpeople set to undergo AV fistula creation Exclusionnot reported

Placebo Type + dosevonapanitaseafter AV fistula creation dropwise on exposed artery, anastomosis and vein

Duration10 min, immediately after fistula creation

Unclear

CommentAbstractrandomised

Unclear

CommentNot reported









Yes

Commentmatching placebo

Yes

Commentdouble-blind

Unclear

CommentNot reported

Yes 0% No, adverse events not reported

Evaluated for Inclusion? Not reportedEnrolled? 313Included Comparator Group? Not reportedIncluded Group 1? Not reported2:1 randomisation

AgeNot reported

Male Sex %Not reported

Not reported

OutcomeDefined As






Effect Measure


Certainty of the evidence

Primary unassisted patencytime to first thrombosis or intervention to restore flow

1 year 31% 42% Not reported Not reported Not reported Not reported

0.25 Unclear

Secondary patencytime to abandonment

1 year 74% 61% Not reported Not reported Not reported Not reported

0.048 Unclear

of 226





KimEffectiveness of beraprost sodium in maintaining vascular access patency in patients on hemodialysis2017International Urology and Nephrology[96]


1South-Korea

2013-2014 Inclusion: maintenance haemodialysis ≥3 months; previous vascular access failure, undergoing new placement AVF or AVGExclusion:history of bleeding; concurrent warfarin or antiplatelet

TypeNot reported

DoseNot reported

DurationNot reported

TypeBeraprost sodium

Dose120 µg/day

Duration2 years

Enrolled: 140Randomised: 55Included Comparator Group: 32Included Intervention Group: 23Analysed:

AgeMale Sex %








Agecontrol 66 ±11; intervention 66 ±10 yearsMale Sex %control group 60%; intervention group: 70%

AVF: n=41AVG: n=14

UnclearComment'patient were randomly assigned', no further details


NoComment'not treated in a blinded manner'

NoCommentopen-lable

UnclearCommentprobably not; not stated


The authors declare no conflict of interest or financial support






Effect Measure


Evidence Certainty


Thrombosis-free and intervention-free access survival

2 years 83% 38% Not reported Not reported Not reported

Not reported

p=0.01 Low - high risk of bias due to control group being unclear, likely not placebo, randomisation procedure unclear, lack of blinding. In addition, no protocol, unclear whether primary outcome not hindsight selection; insufficiently detailed numeric outcome reporting; outcome corresponds to mean survival time of 1.8 versus 1.6 years


Thrombosis-free and intervention-free access survival


Not reported

p=0.46 Low - high risk of bias due to control group being unclear, likely not placebo, randomisation procedure unclear, lack of blinding. In addition, no protocol, unclear whether primary outcome not hindsight selection; insufficiently detailed numeric outcome reporting.

Severe bleeding

not stated Not reported

Not reported Not reported Not reported Not reported Not reported

No severe bleeding in any group'

Not reported

Low - insufficient details

of 226

Chapter 9. Cannulation techniques for AV fistulasAuthorYearJournal

Study Design Patient Characteristics Amstar Score N RCTs included

Grudzinski2013Seminars in Dialysis[99]

Systematic review Inclusion: clinical trials, cohort studies, case seriesExclusion: studies without original or usable data or patient-important outcomes

4/11 2 of 5 RCTs

Results: Narrative summary only


Study Design Patient Characteristics Comparator - Type

Intervention - Type

Amstar Score

N RCTs included

N Patients included

Patients

WongButtonhole versus rope-ladder cannulation of arteriovenous fistulas for hemodialysis: a systematic review2014American Journal of Kidney Diseases[100]

Systematic Review InclusionRCTS (and observational) studies comparing buttonhole versus rope ladder cannulation in incident or prevalent AV fistulas in adult home or in-centre dialysis patients Exclusionabstracts, case reports, not original research, outcomes not considered by review, non-English publications

Rope ladder or usual care

Buttonhole 8/11 5 of 5 RCTs

403 Agerange 49 years -70 years

Male Sex %range 38%-82%

OutcomeDefined As

Study Time point of the Outcome








P-Value

Evidence Certainty

Access failureAccess survival at 1 year usable for successful dialysis

Vaux 2013

1 year 9 0 69 58 log rank Not reported

Not reported

Not reported

0.01 Low, reported in 2 of 5 studies only

Access survival time until of access abandonment, death, modality transfer or end of F/U

MacRae 2014

2 years 16 months (Interquartile range 10.6-29.3)

18.4 months (Interquartile range 10.9-32.7)

0.2

Infectionheterogeneous definition in original articles

Up to 1 year

3 18 162 173 Relative Risk 5.0 1.65 15.14 Moderate, although heterogeneous definition and reported only in 4 of 5 trials

Pain different scales used in original studies

Standardised Mean Difference

0.34 0.76 1.43 Moderate

of 226


Study Design Patient Characteristics Comparator - Type

Intervention - Type

Amstar Score

N RCTs included

N Patients included

Patients

MuirButtonhole cannulation and clinical outcomes in a home hemodialysis cohort and systematic review2014Clinical Journal of the American Society of Nephrology[101]

Systematic Review InclusionRCTS (and observational) studies comparing buttonhole versus rope ladder cannulation in incident or prevalent fistulas in adult home or in-centre dialysis patients Exclusionnot reported

Rope ladder or usual care

Buttonhole 6/11 4 406 Agenot reported

Male Sex %not reported

OutcomeDefined As

Study Time point of the Outcome





Effect Measure





Infection heterogeneous definition in original articles

Struthers 2010 [104]Chow 2011 [105]MacRae 2012 [106]Vaux 2013 [102]

Not reported


3.34 0.91 12.20 Not reported

ModerateHeterogeneous outcome definition and very wide confidence interval

AuthorYear

CharacteristicsIn centre/home

Control cannulation technique

Outcome Follow-up Outcome valueButtonhole

Outcome value Control

Analysed Participants Buttonhole

Analysed Participants Control

P-value N enrolled

Toma 2003 [103] Cannulation by the nurse

Not reported Erythema at the puncture site

3 months 1 0 37 43 86

Struthers 2010 [104] Hospital and satellite units

Rope ladder Fistula infection 6 months 1 0 22 25 56

Vaux 2013 [102] In-centre Rope ladder + area Bacteraemia 1 year 0 20.09/1000 AV fistula days

58 69 140

Vaux 2013 [102] In-centre Rope ladder + area Exit site infection 1 year 20.12/1000 AV fistula days

0 58 69 140

MacRae 2012 [106] In-centre Not reported Bacteremia 8 weeks 1 0 P=1.00 140MacRae 2012 [106] In centre Not reported Localized infection 8 weeks 50/1000 sessions 22.4/1000 P=0.003 140MacRae 2014 [107] In-centre Not reported Infection 12 months 12 0 70 69 P<0.001 140MacRae 2014 [107] In-centre Not reported Staf aureus bacteremia 12 months 3 0 70 69 140MacRae 2014 [107] In-centre Not reported localized infection 12 months 9 0 70 69 140Chow 2011 [105] In-centre and

homeRope ladder Infection at cannulation

site6 months 4 (or 5) 1 (or 0) 35 35 P=0.1 70

of 226

Chapter 10. Needle typesAuthorTitlePublication YearJournal

Study Design

LocationN Centres




MooreComparison of large-gauge hollow-bore haemodialysis access needles: a randomised controlled trial 2015Nephrology[112]


Australiasingle centre

2015 Inclusionprevalent ESKD patients dialyzed on an AV fistula or AV graftExclusionNot reported

Type + doseStandard needles

Duration6 months

Type + doseSafety needles(Nipro SafeTouch™ needles)

Duration6 months

Not clear

Commentabstract available only

Not clear










Not clear


Not clear


Unclear

CommentNot reported

Not reported Not reported

No Evaluated for Inclusion? Not reportedEnrolled? 39Included Comparator Group? 19Included Group 1? 20

AgeNot reportedMale Sex %Not reported

Not reported

OutcomeDefined As



Outcome Value Intervention Group 1Group 2


Analysed Participants Intervention Group 1 Group2

Effect Measure


EvidenceCertainty

Acute access associated complicationsincluding needle stick injuries, fistula ‘blows’, andneedle dislodgement

6 months Not reported Not reported Not reported Not reported Not reported

Not reported

Not reported

Low

Needle stick injuriesEvents occurred on cannulation and during dialysis

6 months 0 0 19 20 Not reported

Not reported

Not reported

Low

InfiltrationsEvents occurred on cannulation and during dialysis

6 months 24 15 19 20 Not reported

Not reported

Not reported

Low

of 226


Study Design

LocationN Centres




MorselliSharp versus blunt dialysis needle use with buttonhole method: open randomised trial2015Journal of Renal Care[113]


Italysingle centre

2015 Inclusionadults, out-patients with ESKD undergoinghaemodialysis treatment via an AV fistula using the buttonhole methodExclusionusing the buttonhole method during a training phase

Type + doseSharp needles

Duration1 month

This study subdivided the cannulation of the participants’ AV fistula into venous and arterial accesses, followed by randomization according to the type of needle, either sharp or blunt.

Type + doseBlunt needles

Duration1 month

Not clear

CommentNot reported

Low risk

CommentRandomised sequences were concealedin sealed, non-transparent envelopes









Not clear

CommentNot reported

High risk

CommentOpen trial.

Unclear

CommentNot reported

Not reported Not reported

Unclear Evaluated for Inclusion? 60Enrolled? 35Included Comparator Group? 35Included Group 1? 35

AgeNot reportedMale Sex %74

Diabetes nephropathy 6%

OutcomeDefined As



Outcome Value Intervention Group 1Group 2


Analysed Participants Intervention Group 1Group2

Effect Measure


Evidence Certainty

Failed cannulationthe failure of the needle to locate or reach the vessel (events)

1 month 6 24 166 venous accesses166 arterial accesses

169 venous accesses169 arterial accesses

Not reported

Not reported

Not reported

Low

Difficulty of insertiondifficulty with accessing the tunnel, although ultimately reaching the vessel



Not reported

Not reported

Not reported

Low

The incidence of the trampoline effectThe inappropriate placement of the needle that was inserted into the vessel, although ultimately reaching the vessel



Not reported

Not reported

Not reported

Low

Pain at the moment of insertion and during dialysisaccording to the Numerical Rating Scale(Not reported)

1 month 1.94 (S. var. 1.98)-Mean for venous 2.09 (S. var. 2.14) Mean for arterial

1.87 (S. var. 2.23)-Mean for venous 2.07 (S. var. 2.15) Mean for arterial



Not reported

Not reported

Not reported

Low

of 226


Study Design

LocationN Centres





MarticorenaRandomised pilot study to compare metal needles versus plastic cannulae in the development of complications in hemodialysis access2018Journal of Vascular Access[114]


Canada1

2013 - 2015

InclusionIncident and prevalent patients in-centre chronic haemodialysisFunctioning AV fistulaWritten informed consentExclusionrequiring plastic cannula >30 days after first cannulationCognitive or language barrierAcute medical illness that impaired the ability to provide informed consent

2 metal needles; 15G

2 plastic cannulas; 17G25 mm or 33 mm length

Yes

Commentpatients were randomised in 1:1 ratio using a randomization computer-simulated protocol and closed envelopes

Unclear

Commentclosed envelopes - material not reported








No

Commentpatients were not blinded to the device

Yes

Commentnurses may or may not have been, primary nephrologist was blinded, as was the interventional radiologist or surgeon

Unclear

Commentnot reported

Yes 0% Yes Evaluated for Inclusion? 189Enrolled? 33Included Comparator Group? 17Included Group 1? 16

Age62 years +/- 15 years

Male Sex %61%

30 AV fistulas; 3 AV grafts

OutcomeDefined As






Effect Measure


Evidence Certainty

PatencyHaving received a procedure for stenosis, thrombosis, or aneurysm formation


0.65 0.37 to 1.14 Low. Very wide confidence interval spanning both important reduction and increase in risk. Difference in baseline risk with possibility for regression to the mean overestimating treatment effect. Some indirectness in the outcome. mean follow-up intervention group 10,9 months; comparator group 8,9 months.

ComplicationsInfiltration during cannulation or hemodialysis


0.53 0.29 to 0.97 Low. Very small study. Optimal information size not met.

of 226

Chapter 11. Timing of intervention for AV fistula thrombosisAuthorTitlePublication YearJournal

Study design Comparison Outcome definitions Patient characteristics

Results

DiskinThe importance of timing of surgery fir hemodialysis vascular access thrombectomy1997Nephron[117]

Treatment: SurgicalTechnique: Surgical thrombectomy Follow-up duration: max 6yStudy type: retrospective nested case-control study

Average time to intervention for successful vs unsuccessful intervention

Declotted AV access: surgery involved removal of thrombus, without implantation of new PTFE material of change in anatomic construction

Number of AV fistulas: 188Mean age: Not reportedMale: Not reportedTime to intervention: averaged by outcome

Surgery <24 h: similar numbers of declotted vs new AV fistulasSurgery 1-7 days: 20% AV fistulas declottedSurgery >7 days: 10% AV fistulas declotted

Sadaghianloo Early surgical thrombectomy improves salvage of thrombosed vascular accesses2014Journal of Vascular Surgery[118]

Treatment: SurgicalTechnique: Surgical thrombectomy with percutaneous transluminal angioplasty and/or stent placement Follow-up duration: 669 daysStudy type: Retrospective cohort study

Thrombectomy <6h vs >6h after diagnosis of thrombosis

Technical success: re-establishment of a functional access that was used for the following dialysis sessionPrimary patency: interval from the time of the initial thrombectomy procedure (index procedure) until any intervention designed to maintain or re-establish patency, or until access rethrombosisSecondary patency: interval from the time of the index procedure until access abandonment, rethrombosis, or performance of interventions designed to maintain or re-establish functionality in a thrombosed access

Number of AV fistulas: 82 Mean age: 66 years ± 16 yearsMale: 52%Time to intervention: G1: 3.6±1.2h, G2: 10.3±5.4h

Successful declotting AV fistula: 43/51 (84%) vs 17/23 (74%); Relative Risk=1.14; 95%CI 0.87 to 1.49

El-Damanawi Successful restoration of arteriovenous dialysis access patency after late interventionClinical Kidney Journal2015[119]

Treatment: Surgical, endovascular or pharmacologicalTechnique: percutaneous angioplasty, thrombolysis with local tissue plasminogen activator, mechanical thrombectomy or combinationFollow-up duration: 90dStudy type: Retrospective cohort study

Intervention <2 days versus ≥2 days after diagnosis

Success: no immediate post intervention failurePrimary patency: failure at 3 months post-intervention

Number of Accesses: 60Mean age: Not reportedMale: Not reportedTime to intervention: Not reported

Success: G1: 29/41 (71%), G2: 12/19 (89%); Multilevel mixed model immediate failureOdds Ratio 0.55; 95%-CI 0.31 to 0.99Primary patency: 3m: G1: 20/41 (49%), G2: 12/19 (63%)Multilevel mixed model 3-month failureOdds Ratio 0.68; 95%-CI 0.36 to 1.27*Includes 28 people with AV fistulas and 13 people with AV grafts

GraorLocal thrombolysis in the treatment of thrombosed arteries, bypass grafts, and arteriovenous fistulasJournal of Vascular Surgery1985[120]

Treatment: PharmacologicalTechnique: streptokinaseFollow-up duration: not definedStudy type: Retrospective cohort study

Thrombolysis >4 days vs ≥4 days after diagnosis

Clinical success: angiographically documented dissolution of the thrombus, symptomatic improvement of the patient's claudication or ischemia, and improvement in the non-invasive hemodynamic studies (study included not only AV fistula accesses, however only AV fistulas are reported here)

Number of AV fistulas: 27Mean age: not definedMale: not definedTime to intervention: average time per group not reported

Successful declotting AV fistula: G1: 16/19 (84%), G2: 3/8 (38%); Relative Risk 2.25; 95%CI 0.90 to 5.61

of 226

Chapter 12. Surgical and endovascular interventions for AV access thrombosisAuthorTitlePublication YearJournal





UflackerThrombosed dialysis access grafts: randomised comparison of the Amplatz thrombectomy device and surgical thrombo-embolectomy 2004European Radiology[127]



1994 to 1997 Inclusion: Recent (within 1-week) thrombosis of vascular access or of native arteries or veins related to vascular accessExclusion: Acute or chronic thrombophlebitis in the respective limb; occlusion or thrombosis of the subclavian artery and branches; active infection in other organs; uncontrolled hypertension

Type: conventional surgical thrombo-embolectomy

Type: thrombectomy using the Amplatz thrombectomy device

Yes

Comment: Patients were randomised by blind computer selection to either treatment.

Unclear










No

Comment: Due to the nature of the intervention, it was not possible to blind the patient, or personnel or study team at the time of the intervention

No Unclear


No 8% Yes Unclear

Comment: financial support and conflict of interest not declared

Evaluated for Inclusion: not reportedEnrolled: 174Actually started: 174Included Comparator Group: 65Included Intervention 1: 109

Not reported

57%

Outcome Outcome Defined As Time Point of the Outcome






Judgement Comments

Immediate thrombectomy success

>90% thrombus removal and ability to dialyze

Same day of salvage procedure

77% 79% 74 140 0.73 Moderate, there are limitations

Subjective outcome definition for allowed size of residual thrombus and unit of analysis is number of interventions performed not number of patients

Patency Presence of pulse, thrill and dialyticfunction

3 months 68% 75% 59 101 0.2 High Unit of analysis is number of patients

Adverse event Any adverse event Not reported 17 events 31 events Not reported Not reported Not reported Moderate, there are limitations

Unit of analysis unclear, details on adverse events not provided

Major adverse events

Formation of a pseudoaneurysm and arterial embolism

Not reported 2 events 2 events Not reported Not reported Not reported High

of 226







BarthHydrodynamic Thrombectomy System versus Pulse-Spray Thrombolysis for Thrombosed Hemodialysis Grafts: A Multicenter Prospective Randomised Comparison 2000Radiology[128]



not reported Inclusion: occluded prosthetic graft for < 14 days without infection, eligibility for lytic therapyEx clusion: not reported

Type: percutaneous transluminal angioplasty with pulse-spray thrombolysisDose: combination of 250,000 Urokinase and 5,000 U of heparin sodium sprayed in 0.2-mL increments every 30 seconds during forward movement of the catheter

Type: percutaneous transluminal angioplasty with thrombectomy by hydrodynamic thrombectomy system

Unclear


Unclear









Male Sex %

No

Comment: Due to the nature of the intervention, it was not possible to blind the patient, or personnel or study team at the time of the intervention

No Unclear


Yes 0% No Yes

Comment: The study was supported by a grant from Boston Scientific. A co-author is an employee of Boston Scientific, which developed, manufactures, and distributes the Oasis Thrombectomy System.


Age: 56 years +/- 15 years

Male Sex %: 48%






P-Value

Evidence Certainty

Judgement Comments

Clinical success

<20% residual thrombus following treatment with assigned therapy and ability to dialyze after treatment

Not reported 81% 89% 58 62 0.24 Moderate, there are limitations

But subjective outcome definition for allowed size of residual thrombus

Patency Graft patent, no graft intervention required 3 months 41% 40% 58 62 0.91 High

of 226


Study Design





VogelThrombosed Hemodialysis Grafts: Lyse and Wait with Tissue Plasminogen Activator or Urokinase Compared to Mechanical Thrombolysis with the Arrow-Trerotola Percutaneous Thrombolytic Device2001Journal of Vascular and Interventional Radiology[129]



2000-2000

Inclusion: patients with thrombosed upper extremity synthetic grafts for less than 2 weeks Ex clusion: graft type other than polytetrafluoroethylene, graft pseudoaneurysm, suspected graft infection

Type: percutaneous transluminal angioplasty with Lyse and Wait technique with tissue plasminogen activatorDose: 4 mg Recombinant tissue plasminogen activator (Alteplase) instilled over the course of 2 minutes while the arterial and venous ends of the graft were gently compressed

Type: percutaneous transluminal angioplasty with Arrow-Trerotola Percutaneous Thrombolytic Device

Unclear


Yes

Comment: Sealed envelopes directingenrolment controlled randomization.









NoComment: Due to the nature of the intervention, it was not possible to blind the patient, or personnel or study team at the time of the intervention

No Unclear


Yes Not reported No UnclearComment: financial support and conflict of interest not declared


62y 55%







Judgement Comments

Anatomic success

Angiographic restoration of flow with 30% residual stenosis in all areas in which angioplasty was attempted

Immediately at intervention

95% 95% 20 20 0.24 Moderate, there are limitations

But subjective outcome definition for allowed size of residual thrombus

Clinical success

Maintenance of graft function for at least one adequatehemodialysis session,

Shortly after intervention

90% 95% Not reported Not reported 0.91 Moderate, there are limitations

Number of patients in whom outcome was assessed not reported

of 226

Supplement 7| Internal Review

1. Standardised internal review form

of 226

of 226

Supplement 8| External ReviewA SurveyMonkey questionnaire was sent to all ERA-EDTA members on 11 January 2019. Furthermore, the survey was sent to the ERA-EDTA National Societies and other organisations with expertise in the area of this guideline and guidelines in general, namely the Canadian Society of Nephrology (CSN), European Society for Vascular Surgery (ESVS), Kidney Disease Improving Global Outcomes (KDIGO), The National Kidney Foundation - Kidney Disease Outcomes Quality Initiative (NKF KDOQI), Kidney Health Australia – Caring for Australasians with Renal Impairment (KHA-CARI), The National Institute for Health and Care Excellence (NICE), Grupo Español Multidisciplinar del Acceso Vascular (GEMAV), Japanese Society of Dialysis Access (JSDA), Vascular Access Society of the Americas (VASA), American Society of Diagnostic and Interventional Nephrology (ASDIN). They had the option to provide comments through the same standardised form as provided in Supplement 7 Internal Review until 7 February 2019. All comments and suggestions were then discussed with the guideline development group by e-mail and teleconference.

63 respondents filled in the online questionnaire, of which 45 were individual respondents and 18 represented an organisation. Professions included 54 nephrologists, 3 surgeons, 1 nurse1 patient’s close kin, 1 guideline technical advisor, 1 specialist in infectious diseases and microbiology, 1 medical director, 1 renal technologist/correctional dialysis programme manager. Countries represented are shown in the figure below.

of 226

Green = agree; Yellow = neutral; Red = disagree

of 226

Chapter 1. Medical treatments for promoting AV fistula maturation1.1 We suggest aspirin, ticlopidine, or clopidogrel, given in adults with end-stage kidney disease

during the first two months after AV fistula creation for the sole purpose of improving maturation, reduces thrombosis, but has uncertain effects on maturation and bleeding. (2C)

External review results: 11/20 agreed; 2/20 were neutral; 4/20 disagreed but left no comment; 3/20 disagreed and provided a comment.

The statement is not entirely conform GRADE, which should either recommend or suggest for or against an intervention. Evidence does not show improved maturation, only reduction in thrombosis; the statement should reflect that. It is ambiguous in the way it is currently written.

The statement is interpreted as supporting treatment and the absence of evidence on bleeding risk is insufficiently weighted.

Benefit likely only present in patients at high risk for thrombosis. Other factors such as vessel choice could have biased the results.

Consensus: It is agreed the statement in its current form is not actionable and may lead to it being

misinterpreted as being in favour of the intervention. The guideline development group felt they could not give a discretionary recommendation for or against the interventions, but that it should be made clearer individual decisions needed balancing of the various benefits and harms. The format of the statement was changed such that it would be actionable.

1.2 We suggest perioperative heparin, given in adults with end-stage kidney disease during AV fistula creation, may increase AV fistula patency at one month but comes at a cost of an important increase in bleeding complications. (2C)

External review results: 13/19 agreed; 3/19 were neutral; 3/19 disagreed but left no comment.

The statement is not entirely conform GRADE which should either recommend or suggest for or against; some reviewers interpreted the statement as supportive of heparin.


misinterpreted as being in favour of the intervention. The group felt they could not give a discretionary recommendation for or against the intervention, but that it should be made clearer individual decisions needed balancing of the various benefits and harms. The format of the statement was changed such that it would be actionable.

1.3 We suggest far infrared therapy, applied in adults with end-stage kidney disease during the first three months after AV fistula creation may reduce thrombosis, but has uncertain effects on maturation and bleeding. (2C)

External review results: 4/18 agreed; 8/18 were neutral; 5/18 disagreed but left no comment; 1/18 disagreed and left a comment.

The statement is not conform GRADE which should either recommend or suggest for or against; some reviewers interpreted the statement as for or against FIR; most do not seem to hold a strong opinion.

Not used, limited evidence; another feels it is highly effective and cost-effective.


misinterpreted as being in favour of the intervention. The group felt they could not give a discretionary recommendation for or against the interventions, but that it should be made clearer

of 226

individual decisions needed balancing of the various benefits and harms. The format of the statement was changed such that it would be actionable.

1.4 There are insufficient RCT data to make a recommendation for ticagrelor, prasugrel, dipyridamole, sulphinpyrazone, warfarin or other oral anticoagulants, fish oil, vonapanitase, glycerine trinitrate, iontophorectic injection of Salvia miltiorrhiza, or prednisolone for improving AV fistula maturation in adults with end-stage kidney disease. (-D)

External review results: 13/17 agreed; 3/17 were neutral; 1/17 disagreed but left no comment.

Additional RCT on Beraprost sodium identified by external reviewer (Kim et al). Statins to be added to the list of treatments with insufficient RCT data as was identified in the

PICO. Additional RCT on Beraprost sodium identified by external reviewer not to be included here, but

in chapter 8 as it did not assess maturation, but patency at two years. Advice for clinical practice to not stop antiplatelet treatment is very helpful as the previous

statements are not actually recommendations. Could this be taken up in the summary recommendations?

Consensus: Statins were added to the list of treatments with insufficient data. The advice for clinical practice was taken up in the summary of recommendations

Chapter 2. Surgical and endovascular interventions for promoting AV fistula maturation

2.1 We suggest using regional block anaesthesia rather than local anaesthesia for AV fistula creation in adults with end-stage kidney disease. (2C)

External review results: 3/10 agreed; 5/10 were neutral; 0/10 disagreed but provided no comment; 2/10 disagreed and left comment.

Not taken into account: regional block anaesthesia may complicate procedure, increase cost, delay the access creation procedure.

Main advantage of regional block anaesthesia is vein dilation, which can be achieved also with warm conditions.

Consensus The evidence to recommendation process did take into account the fact that a switch to regional

block anaesthesia may complicate the procedure, increase cost and possibly even delay the procedure. It contributed to the recommendation being discretionary or weak, rather than strong. It was agreed this could have been discussed more explicitly in the text. A sentence was added to the rationale reflecting the uncertainties of the effect of switching to regional block anaesthesia on cost and logistics and how this limited the strength of the recommendation issued.

It is agreed it would be helpful to mention in the rationale that inducing venodilation via local heat application could improve outcome with local anaesthesia.

2.2 We suggest there is insufficient evidence to support end-of-vein to side-of-artery over side-of-vein to side-of-artery anastomosis for AV fistula creation in adults with end-stage kidney disease. (2C)

External review results: 6/8 agreed; 1/8 was neutral, 1/8 disagreed and provided a comment.

There can be situations where one technique may be preferred over another from a clinical perspective.

of 226

Consensus: No changes were felt to be indicated.

Chapter 3. Surgical and endovascular interventions for non-maturing AV fistulas

3.1 We suggest there is insufficient evidence to support open surgical over endovascular interventions as the preferred treatment for non-maturing AV fistulas in adults with end-stage kidney disease. (2D)

External review results: 6/9 agreed; 2/9 were neutral, 1/9 disagreed and provided a comment.

Comments are related to what is done in practice, but no evidence-based justification offered.



4.1 We suggest a standardised exercise program involving hand and wrist motion may improve AV fistula maturation in adults with end-stage kidney disease. (2C)

External review results: 6/9 agreed; 1/9 was neutral; 1/9 disagreed but provided no comment; 1/9 disagreed and left a comment.

Comment related to their clinical experience not being in line with the evidence. One reviewer suggested to change ‘hand and wrist motion’ to ‘hand and arm exercises’.

Consensus: The group agreed that the reviewer suggested description better reflected the intended meaning

and so ‘hand and wrist motion’ was changed to ‘hand and arm exercises’.

4.2 There is insufficient evidence to support specific exercise programmes or physical interventions to promote AV fistula maturation in adults with end-stage kidney disease. (-D)

External review results: 4/8 agreed; 2/8 were neutral; 1/8 disagreed but provided no comment; 1/8 disagreed and left a comment.

Invalid comment stating evidence for exercise programmes is accumulating. Indeed, this is why the first recommendation suggests in favour of a programme. This particular statement says there is insufficient evidence to prefer one over another.



5.1 We recommend giving preoperative antibiotic prophylaxis for AV graft insertion in adults with end-stage kidney disease. (1C)

External review results: 12/12 agreed.

Consensus: No changes needed.

5.2 We suggest giving preoperative antibiotic prophylaxis for complex AV access procedures in

of 226

adults with end-stage kidney disease. (2D)


Three comments related to the 'complexity' not being clear. Perhaps it would have been better to state the third guideline first (i.e. no antibiotics for simple AVF) and then have this guideline for 'other' rather than 'complex' AV access procedures.

Consensus: The advice for clinical practice was added to the summary recommendations. Definitions have

been added there. The group felt it to be better to start with the strongest recommendation and the highest certainty

to underpin the statement. Hence, the order was not changed.

5.3 We suggest not giving preoperative antibiotic prophylaxis for simple AV access procedures in adults with end-stage kidney disease. (2D)


One comment about antibiotic prophylaxis being necessary in all cases, but that the type of antibiotic differs. However, no evidence-based rationale provided.

Consensus: The advice for clinical practice was added to the summary recommendations. Definitions have

been added there.


AV Fistulas

6.1 In adults requiring haemodialysis, we suggest AV fistulas can be cannulated four weeks after creation if they are considered suitable for cannulation on clinical examination. (2C)


All comments relate to different cut-offs being used by reviewers. However, no evidence-based argument provided.


6.2 In adults requiring haemodialysis, we recommend against cannulating AV fistulas sooner than two weeks after their creation. (1B)

External review results: 19/21 agreed; 1/21 was neutral; 1/21 disagreed and provided a comment.

Comment relates to the reviewer being against one size fits all as all AVFs mature differently.


6.3 In adults requiring haemodialysis, we suggest against cannulating AV fistulas in between two

of 226

and four weeks after their creation, unless this can avoid placement of a central venous catheter for haemodialysis. (2C)

External review results: 19/21 agreed; 2/21 were neutral; 2/21 disagreed and provided a comment.

Comment relates to the time-frame; clinical practice suggests cannulating after 3 weeks is possible without complications. However, no numerical data provided.


AV Grafts

6.4 In adults requiring haemodialysis, we recommend ‘early cannulation type’ AV grafts can be cannulated as soon as wound healing permits. (1B)

External review results: 15/21 agreed; 4/21 were neutral; 2/21 disagreed but provided no comment.


6.5 In adults requiring haemodialysis, we suggest against cannulating a ‘standard type’ AV graft sooner than two weeks after insertion, unless this can avoid placement of a central venous catheter for haemodialysis. (2B)




AV Grafts7.1 We suggest against technical surveillance in addition to clinical monitoring of a functional AV

graft to detect and pre-emptively correct a haemodynamically important AV access stenosis in adults. (2C)


Current practice and EBPG guidelines so far emphasized the use of standardized monitoring of access flow with various techniques. More specifically, ultrasound and color Doppler technique offer an easy, relatively inexpensive and non-invasive tool for assessing access flow periodically, which can promptly identify complications and set alarm for interventions.

Recommendation can very likely not be issued because disparate data and centre policies exist and individual approach would possibly be the best.

Studies have shown that regular physical examination is both sensitive and specific for identifying vascular access issues (Asif et al). Moreover, in our clinical practice with significant load of patients, the possibility of performing technical surveillance for all patients is questionable. Apart from the cost constraints, the assessment of benefits (especially long-term AV patency) weighed against the risks / harms does not suggest any significant additional utility of technical surveillance in this setting.

In our experience, as an emergency hospital with both nephrology-dialysis department and vascular surgery department and many presentations for vascular access complications, interventions for AVF graft complications are risky and we provide them only if completely

of 226

justified. Our current practice is to technically investigate an AVF graft only if there are clinical/physical modifications.

Technical surveillance is usually also performed in AV grafts, but I am aware it is not evidence-based.

Consensus: The guideline development group agreed the topic would remain a contentious one. First, much

relies on whether or not one feels the subgroup analysis (AV fistulas versus AV grafts) is valid. Second, the studies included in the evidence base used different techniques for identifying a haemodynamically important stenosis and for subsequently pre-emptively correcting it. However, it needs to be emphasized that the systematic review looked for heterogeneity related to technique (both visually and statistically) and did not find any. Subsequently, most members in the guideline development group agreed the evidence to date did not show a clear enough benefit to support a recommendation for routine use of AV graft monitoring at this point. In response to the external review process and ongoing discussions within the group they agreed, however, that adding a clause for monitoring within a research setting would lift some of the contentiousness and leave some room for encouraging research in this field. Hence, the recommendation was changed to: ‘We suggest technical surveillance in addition to clinical monitoring of a functional AV graft to detect and pre-emptively correct a haemodynamically important AV access stenosis in adults only occurs in the context of a clinical study. (2C)’

AV Fistulas7.2 We suggest the evidence for technical surveillance in addition to clinical monitoring of a

functional AV fistula to detect and pre-emptively correct a haemodynamically important AV access stenosis in adults is inconclusive and needs more research. (2C)


In our experience as emergency nephrology-dialysis department with high addressability, the large number of acute native AVF losses, requiring temporary vascular access (with known complications), represents a strong recommendation for doppler ultrasound surveillance added to clinical surveillance. Clinical experience of the nephrologists may differ, dialysis centres are heterogenous and we must set a rule to ensure the preservation of the AVF quality. We do not consider "additional demands on radiology services" or the infectious risk as reasons for renouncing at a pre-emptive correction. Furthermore, we strongly advise technical surveillance for diagnosing high-flux AVF - a complication too little discussed

Contrary to AVF grafts, AV fistula dysfunction gives early signs, especially during the HD session (blood line pressure monitors)

There is evidence that technical surveillance has no value in predicting stenoses and pre-emptive intervention will not improve outcome!

I agree that the evidence for technical surveillance is inconclusive even if it is part of our daily practice.

Consensus: The guideline development group accepts the current evidence is appreciated and weighted

differently by various stakeholders. Precisely because of the difficulty in attaining consensus within the group, the guideline development group ultimately refrained from speaking out for or against technical surveillance. The results from the external review process largely seemed to support that stance, such that no further changes were made to the statement at this point.

Chapter 8. Medical treatments for maintaining long term AV access patency

AV Fistulas

of 226

8.1 We suggest clopidogrel, given to adults with end-stage kidney disease in the six months following AV fistula creation may reduce primary failure by thrombosis, but comes at an unknown risk of bleeding. (2C)


Long term patency should be defined as 3-5 years, not 6 or 12 months. Why advise clopidogrel and not aspirin? Insufficient stress on bleeding risk. Validity/Reliability – the evidence being used to suggest that clopidogrel may reduce primary

fistula failure at one year is limited by poor external validity (Selection bias). The other identified data on clopidogrel was in an RCT that required early termination due to adverse events. Therefore, the conclusion to suggest that clopidogrel may reduce primary failure with unknown risk of bleeding is accurate but should result in a finding of insufficient data to make a clear recommendation. Feasibility - There is certainly flexibility in individual clinician decisions.


misinterpreted as being in favour of the intervention. Given the level of disagreement with the content of the statement, the guideline development group rediscussed the evidence in detail and ultimately agreed it would best be catalogued among the treatments with insufficient evidence for formulating a recommendation.

8.2 We suggest fish oil, given to adults with end-stage kidney disease in the year after AV fistula creation, may improve patency at one year, but comes at an unknown risk of bleeding. (2C)


Insufficient evidence. Unclear – not a recommendation – not conform to GRADE. Not implementable in many countries as not used. Validity/Reliability – the evidence being used to suggest that fish oil may reduce primary fistula

failure, appears to support this recommendation. The limited data on severity of adverse events limits recommending this for most/all patients.


misinterpreted as being in favour of the intervention. The group felt they could not give a discretionary recommendation for or against the interventions, but that it should be made clearer individual decisions needed balancing of the various benefits and harms. The format of the statement was changed such that it would become actionable.

8.3 We suggest far infrared therapy can be considered for improving long term AV fistula patency in adults with end-stage kidney disease. (2C)


Not available. Unclear statement. Evidence inadequate, limited external validity. Feasibility would have to be explored.

of 226

Consensus: The guideline group carefully assessed the comments from the external review process but felt it

would be difficult to formulate a statement that would clarify its meaning further. Hence, no changes were made.

8.4 There are insufficient RCT data to make a recommendation for aspirin, ticlopidine, sulphinpyrazone, warfarin, vonapanitase, cholecalciferol, dipyridamole or dipyridamole combined with aspirin to be given for maintaining long-term AV fistula patency in adults with end-stage kidney disease. (-D)

External review results: 6/11 agreed; 5/11 neutral were neutral to the statement.

Additional RCT on Beraprost sodium identified by external reviewer (Kim et al).

Consensus: An additional RCT on beraprost sodium assessing two- and one-year patency in AV fistulas and

grafts was identified by an external reviewer (Kim et al). The evidence was insufficient to make a recommendation for or against the intervention. It was agreed it should be inserted here.

Statins to be added to the list of treatments with insufficient RCT data as was identified in the PICO.

8.5 We recommend against using warfarin in combination with antiplatelet agents, and the combination of clopidogrel plus high dose aspirin for reducing AV graft thrombosis in adults with end-stage kidney disease. (1C)


Consensus: No changes were felt to be indicated

AV Grafts

8.6 We suggest fish oil, given in the year following AV graft creation in adults with end-stage kidney disease, may improve graft patency at one year, but comes at an unknown risk of bleeding. (2C)


Fish oil in not used routinely in clinical practice in many countries and therefore, in my opinion should not be suggested by some experts.

The limited information on safety of this intervention limits the ability to recommend broadly.

Consensus: It was agreed the statement was not actionable and could lead to it being misinterpreted as being in

favour of the intervention. The group felt they could not give a discretionary recommendation for or against the interventions, but that it should be made clearer individual decisions needed balancing of the various benefits and harms. The format of the statement was changed such that it would become actionable.

8.7 There are insufficient RCT data to make a recommendation for aspirin, ticlopidine, sulphinpyrazone, warfarin, vonapanitase, cholecalciferol, dipyridamole or dipyridamole combined with aspirin to be given for maintaining long-term AV fistula patency in adults with

of 226

end-stage kidney disease. (-D)


The role of medical treatment could be done in one chapter. Additional RCT on Beraprost sodium identified by external reviewer (Kim et al).

Consensus: The choice to separate the two chapters was done purposefully because the processes that govern

maturation and long-term patency were deemed to be different with differential effects of treatments. No changes were felt to be indicated.

An additional RCT on beraprost sodium assessing two- and one-year patency in AV fistulas and grafts was identified by an external reviewer (Kim et al). The evidence was insufficient to make a recommendation for or against the intervention. It was agreed it should be inserted here.

Statins to be added to the list of treatments with insufficient RCT data as was identified in the PICO.


9.1 We suggest against using area technique for cannulating AV fistulas in adults treated with haemodialysis. (2D)

External review results: 10/11 agreed; 1/11 was neutral.

To use rope-ladder technique long segment of AVF is needed. In real life, area technique is more commonly used (see paper of Parisotto).


9.2 We suggest using either a rope ladder or buttonhole technique for cannulating AV fistulas in adults treated with haemodialysis, and letting the choice be dependent on local expertise and AV fistula characteristics. (2D)

External review results: 7/11 agreed; 1/11 was neutral; 3/11 disagreed and provided a comment.

There is no evidence. But probably we should suggest preferring a rope ladder technique when the fistula is long enough and buttonhole only in selected patients.

I practice the buttonhole technique when applicable and have great success with it. Emphasize the infection risk of buttonhole. I do not agree with the button hole technique. The higher risk of infection should be mentioned. I

do totally agree with the rope laser technique Given the very high risk of infection with buttonhole cannulation it should not be recommended

for cannulation routinely.

Consensus: All comments were carefully considered. The advice for clinical practice states that ‘antiseptic

measures and practical aspects of the cannulation procedure are important in reducing the infection risk associated with buttonhole cannulation.’ It was felt that the concerns voiced by external reviewers could remedied by reading the advice for clinical practice in conjunction with the statement. Hence the summary was altered to accommodate this.

Chapter 10. Needle types for AV fistulas

10.1 We suggest using either sharp needles or plastic cannulas for cannulating AV fistulas in adults

of 226

treated with haemodialysis. (2C)

External review results: 9/10 agreed; 1/10 was neutral.


10.2 We recommend using blunt needles only for buttonhole cannulation of AV fistulas in adults treated with haemodialysis. (1D)


Nobody is ever going to use a blunt needle to penetrate the intact skin and therefore there are no data about this procedure. I recommend changing the guideline to: There is no evidence to prefer blunt needles over sharp needles for buttonhole cannulation

Blunt needles were designed for button hole cannulation. Using them for rope ladder technique would probably be very painful.

It should also be said that only blunt needles should be used for the button hole technique! Blunt needles must only be used for established BH sites in AVF

Consensus: All comments were carefully considered. This statement needs to be considered together with the

previous recommendation in this chapter. If we suggest using either sharp needles or plastic cannulas for cannulating AV fistulas, then this holds for all AV fistulas, also those cannulated using buttonhole technique. One may argue the second statement to be redundant, but it was decided to leave it in to emphasize the underlying clinical questions.


11.1 We suggest attempting to declot a thrombosed AV fistula in adults as soon as possible under good conditions and before the next haemodialysis treatment. (2D)

External review results: 13/13 agreed.


11.2 We suggest attempting to declot a thrombosed AV fistula in adults, even if there has been a delay of days to weeks. (2D)

External review results: 5/11 agreed; 4/11 were neutral; 1/11 disagreed but provided no comment; 1/11 disagreed and left a comment

A very uncertain situation and there should be a highly individual approach depending on the patient's condition.

In my centre we prefer to do proximal reanastomosis in case of thrombosis. It means that AVF could be re-established even many years from thrombosis.

Since the success rate differs from cases, the recommendation in GL is questionable.

Consensus: All the comments were carefully considered. The statement does not imply weeks of delay are

acceptable, it states that an attempt should be undertaken, even if for some reason – e.g. the patient has not showed up for dialysis for ten days – the attempt could not occur earlier. This is explained extensively in the rationale. It was felt the comments voiced could not easily be translated into an evidence-based recommendation. A specific systematic review would have to be performed to

of 226

explore the comparison of declotting versus creating a new anastomosis proximal of the thrombosis in case of stenosis. Although this could be an important question to consider for a future update, it was considered beyond the remit of the current guideline.


12.1 We suggest the choice between surgical and endovascular interventions for AV access thrombosis be defined by the condition of the patient and their vascular access, as well as local expertise, as there is no evidence one approach improves outcomes more than any other. (2B)

External review results: 8/10 agreed; 2/10 were neutral.


of 226

24. References

1. Evangelidis N, Tong A, Manns B, et al. Developing a Set of Core Outcomes for Trials in Hemodialysis: An International Delphi Survey. American journal of kidney diseases 2017;70(4):464-4752. Tordoir J, Canaud B, Haage P, et al. EBPG on Vascular Access. Nephrology Dialysis Transplantation 2007;22(suppl_2):ii88-ii1173. Zoccali C, Abramowicz D, Cannata-Andia JB, et al. European best practice quo vadis? From European best practice guidelines (EBPG) to European renal best practice (ERBP). Nephrology Dialysis Transplantation 2008;23(7):2162-21664. Viecelli AK, Tong A, O’Lone E, et al. Report of the Standardized Outcomes in Nephrology–Hemodialysis (SONG-HD) Consensus Workshop on Establishing a Core Outcome Measure for Hemodialysis Vascular Access. American Journal of Kidney Diseases 2018;71(5):690-7005. Van Der Veer SN, Haller MC, Pittens CACM, et al. Setting priorities for optimizing vascular access decision making - An international survey of patients and clinicians. PLoS ONE 2015;10(7)6. Cross NB, Craig JC, Webster AC. Asking the right question and finding the right answers. Nephrology 2010;15(1):8-117. Shea BJ, Bouter LM, Peterson J, et al. External Validation of a Measurement Tool to Assess Systematic Reviews (AMSTAR). PLOS ONE 2007;2(12):e13508. Higgins JPT, Altman DG, Gøtzsche PC, et al. The Cochrane Collaboration’s tool for assessing risk of bias in randomised trials. BMJ 2011;343:d59289. Wells G, Shea B, O'Connell D, et al. The Newcastle–Ottawa Scale (NOS) for Assessing the Quality of Non-Randomized Studies in Meta-Analysis. http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp.10. Whiting PF, Rutjes AS, Westwood ME, et al. Quadas-2: A revised tool for the quality assessment of diagnostic accuracy studies. Annals of Internal Medicine 2011;155(8):529-53611. Guyatt GH, Oxman AD, Vist GE, et al. GRADE: an emerging consensus on rating quality of evidence and strength of recommendations. BMJ 2008;336(7650):92412. Palmer SC, Di Micco L, Razavian M, et al. Antiplatelet therapy to prevent hemodialysis vascular access failure: systematic review and meta-analysis. American Journal of Kidney Diseases 2013:112-12213. Tanner NC, Da Silva A. Medical adjuvant treatment to increase patency of arteriovenous fistulae and grafts. The Cochrane database of systematic reviews 2015(7):CD00278614. Palmer SC, Di Micco L, Razavian M, et al. Antiplatelet agents for chronic kidney disease. Cochrane Database of Systematic Reviews 2013:CD00883415. Viecelli AK, Irish AB, Polkinghorne KR, et al. Omega-3 polyunsaturated fatty acid supplementation to prevent arteriovenous fistula and graft failure: a systematic review and meta-analysis of randomized controlled trials. American Journal of Kidney Diseases 2018;72(1):50-6116. Bashar K, Healy D, Browne LD, et al. Role of far infra-red therapy in dialysis arterio-venous fistula maturation and survival: Systematic review and meta-analysis. PLoS ONE 2014;9(8):e10493117. Wan Q, Yang S, Li L, et al. Effects of far infrared therapy on arteriovenous fistulas in hemodialysis patients: a meta-analysis. Renal failure 2017;39(1):613-62218. Smith GE, Souroullos P, Cayton T, et al. A systematic review and meta-analysis of systemic intraoperative anticoagulation during arteriovenous access formation for dialysis. Journal of Vascular Access 2016;17(1):1-519. Harter HR, Burch JW, Majerus PW, et al. Prevention of thrombosis in patients on hemodialysis by low-dose aspirin. New England Journal of Medicine 1979;301(11):577-57920. Dixon BS, Beck GJ, Vazquez MA, et al. Effect of dipyridamole plus aspirin on hemodialysis graft patency. New England Journal of Medicine 2009;360(21):2191-220121. Middleton D, Deichsel G. The prophylaxis of thrombosis in new arteriovenous dialysis shunts in the arm by low dose acetylsalicylic acid and dipyridamole. Boehringer Ingelheim. Internal Report 1992, Placed Published: 1992.

of 226

22. Taber T, Maikranz P, Haag B, et al. Hemodialysis vascular graft stenosis may be altered by lowmolecular weight dextran (LMD), but not by aspirin (ASA). Journal of the American Society of Nephrology 1992;3(3):39723. Abacilar AF, Atalay H, Dogan OF. Oral prostacycline analog and clopidogrel combination provides early maturation and long-term survival after arteriovenous fistula creation: A randomized controlled study. Indian journal of nephrology 2015;25(3):136-14224. Bleyer A, Kidd K, Wilson S, et al. (2017) The effect of topically applied vonapanitase on fistula outcomes: results of the patency-1 trial. National Kidney Foundation 2017 spring clinical meetings late breaking abstracts. American Journal of Kidney Diseases, 10.1053/j.ajkd.2017.03.005.25. Wasse H, Huang R, Long Q, et al. Very high-dose cholecalciferol and arteriovenous fistula maturation in ESRD: a randomized, double-blind, placebo-controlled pilot study. Journal of Vascular Access 2014;15(2):88-9426. Field M, McGrogan D, Marie Y, et al. Randomized clinical trial of the use of glyceryl trinitrate patches to aid arteriovenous fistula maturation. The British Journal of Surgery 2016;103(10):1269-127527. Rongrong W, Li D, Guangqing Z, et al. The effects of iontophoretic injections of Salvia miltiorrhiza on the maturation of the arteriovenous fistula: a randomized controlled trial. Alternative Therapies in Health and Medicine 2016;22(5):18-2228. Voorzaat BM, van Schaik J, van der Bogt KEA, et al. Improvement of radiocephalic fistula maturation: rationale and design of the Liposomal Prednisolone to Improve Hemodialysis Fistula Maturation (LIPMAT) study - a randomized controlled trial. Journal of Vascular Access 2017;18(Suppl. 1):114-11729. Andrassy K, Malluche H, Bornefeld H, et al. Prevention of p.o. clotting of av. cimino fistulae with acetylsalicyl acid. Results of a prospective double blind study. Klinische Wochenschrift 1974:348-34930. Chen L, Ling YS, Lin CH, et al. Combined use of heparin and anisodamine reduces the risk of early thrombosis in native arteriovenous fistula. Vascular 2013;21(6):369-37431. Irish AB, Viecelli AK, Hawley CM, et al. Effect of Fish Oil Supplementation and Aspirin Use on Arteriovenous Fistula Failure in Patients Requiring Hemodialysis: A Randomized Clinical Trial. JAMA internal medicine 2017;177(2):184-19332. Lewis SR, Pritchard MW, Schofield-Robinson OJ, et al. Continuation versus discontinuation of antiplatelet therapy for bleeding and ischaemic events in adults undergoing non-cardiac surgery. Cochrane Database of Systematic Reviews 2018;7:Cd01258433. Schmidli J, Widmer MK, Basile C, et al. Vascular Access: 2018 Clinical Practice Guidelines of the European Society for Vascular Surgery (ESVS). European Journal of Vascular and Endovascular Surgery 2018;55(6):757-81834. Ibeas J, Roca-Tey R, Vallespin J, et al. Spanish Clinical Guidelines on Vascular Access for Haemodialysis. Guia Clinica Espanola del Acceso Vascular para Hemodialisis. 2017;37 Suppl 1:1-19135. Ismail A, Abushouk AI, Bekhet AH, et al. Regional versus local anesthesia for arteriovenous fistula creation in end-stage renal disease: a systematic review and meta-analysis. Journal of Vascular Access 2017;18(3):177-18436. Bashar K, Medani M, Bashar H, et al. End-To-Side versus Side-To-Side Anastomosis in Upper Limb Arteriovenous Fistula for Dialysis Access: A Systematic Review and a Meta-Analysis. Annals of vascular surgery 2018;47:43-5337. Aitken E, Jackson A, Kearns R, et al. Effect of regional versus local anaesthesia on outcome after arteriovenous fistula creation: a randomised controlled trial. Lancet 2016;388(10049):1067-107438. Meena S, Arya V, Sen I, et al. Ultrasound-guided supraclavicular brachial plexus anaesthesia improves arteriovenous fistula flow characteristics in end-stage renal disease patients. Southern African Journal of Anaesthesia and Analgesia 2015;21(5):12-1539. Shoshiashvili V, Tataradze A, Beglarishvili L, et al. Influence of type of anesthesia on hemodynamic parameters and outcome of dialysis arteriovenous fistula operations. Georgian Medical News 2015;-(249):20-2740. Sahin L, Gul R, Mizrak A, et al. Ultrasound-guided infraclavicular brachial plexus block enhances postoperative blood flow in arteriovenous fistulas. Journal of Vascular Surgery 2011:749-753

of 226

41. Thomsen M, Bengtsson M, Lassvik C, et al. Adjuvant intravenous sympathetic block with guanethidine in construction of arteriovenous fistulas for blood access. Acta Chirurgica Scandinavica 1983;149(2):141-14542. Yildirim V, Doganci S, Yanarates O, et al. Does preemptive stellate ganglion blockage increase the patency of radiocephalic arteriovenous fistula? Scandinavian Cardiovascular Journal 2006;40(6):380-38443. Wedgwood KR, Wiggins PA, Guillou PJ. A prospective study of end-to-side vs. side-to-side arteriovenous fistulas for haemodialysis. British Journal of Surgery 1984;71(8):640-64244. Mozaffar M, Fallah M, Lotfollahzadeh S, et al. Comparison of efficacy of side to side versus end to side arteriovenous fistulae formation in chronic renal failure as a permanent hemodialysis access. Nephro-Urology Monthly 2013;5(3):827-83045. Khan MW, Khan MM, Qadir I, et al. Comparative study of efficacy of end-to-side with side-to-side arteriovenous fistula in patients on hemodialysis. Pakistan Journal of Medical and Health Sciences 2015;9(1):235-23846. Schild AF, Raines J. Preliminary prospective randomized experience with vascular clips in the creation of arteriovenous fistulae for hemodialysis. American Journal of Surgery 1999:33-3747. Walker S. U Clips for arteriovenous anastomosis: a pilot, randomized study. ANZ Journal of Surgery 2012:630-63248. Zeebregts CJ, van den Dungen JJ, van Det RJ, et al. Randomized clinical trial of continuous sutures or non-penetrating clips for radiocephalic arteriovenous fistula. British Journal of Surgery 2004;91(11):1438-144249. Beigi AA, Masoudpour H, Alavi M. The effect of ligation of the distal vein in snuff-box arteriovenous fistula. Saudi Journal of Kidney diseases and Transplantation 2009;20(6):1110-111450. Laskar M, Cornu E, Leman A, et al. Anastomosis of small caliber vessels. Comparison between continuous or interrupted suture. Presse Medicale 1988;17(22):1152-115351. Kakkos SK, Tsolakis IA, Papadoulas SI, et al. Randomized controlled trial comparing primary and staged basilic vein transposition. Frontiers in Surgery 2015;2:1452. Veroux P, Giaquinta A, Tallarita T, et al. Primary balloon angioplasty of small (<2 mm) cephalic veins improves primary patency of arteriovenous fistulae and decreases reintervention rates. Journal of Vascular Surgery 2013;57(1):131-13653. Tordoir JHM, Zonnebeld N, van Loon MM, et al. Surgical and endovascular intervention for dialysis access maturation failure during and after arteriovenous fistula surgery: review of the evidence. European Journal of Vascular and Endovascular Surgery 2018;55(2):240-24854. DerDerian T, Hingorani A, Boniviscage P, et al. Acute complications after balloon-assisted maturation. Annals of Vascular Surgery 2014;28(5):1275-127955. Lee T, Tindni A, Roy-Chaudhury P. Improved cumulative survival in fistulas requiring surgical interventions to promote fistula maturation compared with endovascular interventions. Seminars in Dialysis 2013;26(1):85-8956. Long B, Brichart N, Lermusiaux P, et al. Management of perianastomotic stenosis of direct wrist autogenous radial-cephalic arteriovenous accesses for dialysis. Journal of Vascular Surgery 2011;53(1):108-11457. Harms J, Rangarajan S, Young C, et al. Outcomes of arteriovenous fistulas and grafts with or without intervention before successful use. Journal of Vascular Surgery 2016;64(1):155-16258. Ibeas J, Roca-Tey R, Vallespin J, et al. Spanish Clinical Guidelines on Vascular Access for Haemodialysis. Nefrologia 2017;37 Suppl 1:1-19159. Workgroup VA. Clinical practice guidelines for vascular access. American Journal of Kidney Diseases 2006;48 Suppl 1:S176-24760. Salimi F, Majd NG, Moradi M, et al. Assessment of effects of upper extremity exercise with arm tourniquet on maturity of arteriovenous fistula in hemodialysis patients. Journal of Vascular Access 2013;14(3):239-24461. Kong S, Lee KS, Kim J, et al. The effect of two different hand exercises on grip strength, forearm circumference, and vascular maturation in patients who underwent arteriovenous fistula surgery. Annals of Rehabilitation Medicine 2014;38(5):648-65762. Fontseré N, Mestres G, Yugueros X, et al. Effect of a postoperative exercise program on arteriovenous fistula maturation: a randomized controlled trial. Hemodialysis International

of 226

2016;20(2):306-31463. Desai S, Mitra A, Arkans E, et al. Early application of an intermittent pneumatic compression device is safe and results in proximal arteriovenous fistula enlargement. Journal of Vascular Access 2018:112972981877329564. Bennion R, Hiatt, Jr., Williams R, et al. A randomized, prospective study of perioperative antimicrobial prophylaxis for vascular access surgery. Journal of Cardiovascular Surgery 1985;26(3):270-27465. Zibari G, Gadallah M, Landreneau M, et al. Preoperative vancomycin prophylaxis decreases incidence of postoperative hemodialysis vascular access infections. American Journal of Kidney Diseases 1997;30(3):343-34866. National Institute for Health and Care Excellence. Surgical site infections: prevention and treatment (NICE Guideline CG74). (02/2017; date last accessed).67. Al Shakarchi J, Inston N. Early cannulation grafts for haemodialysis: An updated systematic review. Journal of Vascular Access 2018:112972981877657168. Rayner HC, Pisoni RL, Gillespie BW, et al. Creation, cannulation and survival of arteriovenous fistulae: Data from the dialysis outcomes and practice patterns study. Kidney International 2003;63(1):323-33069. Saran R, Dykstra DM, Pisoni RL, et al. Timing of first cannulation and vascular access failure in haemodialysis: an analysis of practice patterns at dialysis facilities in the DOPPS. Nephrology Dialysis Transplantation 2004:2334-234070. Medkouri G, Aghai R, Anabi A, et al. Analysis of vascular access in hemodialysis patients: a report from a dialysis unit in Casablanca. Saudi journal of kidney diseases and transplantation : an official publication of the Saudi Center for Organ Transplantation, Saudi Arabia 2006;17(4):516-52071. Culp K, Flanigan M, Taylor L, et al. Vascular access thrombosis in new hemodialysis patients. American Journal of Kidney Diseases 1995;26(2):341-34672. Ravani P, Brunori G, Mandolfo S, et al. Cardiovascular Comorbidity and Late Referral Impact Arteriovenous Fistula Survival: A Prospective Multicenter Study. Journal of the American Society of Nephrology 2004;15(1):204-20973. Wilmink T, Hollingworth L, Stevenson T, et al. Is early cannulation of an arteriovenous fistula associated with early failure of the fistula? Journal of Vascular Access 2017;18(Suppl. 1):92-9774. Wilmink T, Powers S, Hollingworth L, et al. Effect of first cannulation time and dialysis machine blood flows on survival of arteriovenous fistulas. Nephrology Dialysis Transplantation 2018;33(5):841-84675. Allon M, Imrey PB, Cheung AK, et al. Relationships Between Clinical Processes and Arteriovenous Fistula Cannulation and Maturation: A Multicenter Prospective Cohort Study. American Journal of Kidney Diseases 2018;71(5):677-68976. Ferring M, Henderson J, Wilmink T. (2014) Accuracy of early postoperative clinical and ultrasound examination of arteriovenous fistulae to predict dialysis use. Journal of vascular access, 10.5301/jva.5000210.77. Jaberi A, Muradali D, Marticorena RM, et al. Arteriovenous fistulas for hemodialysis: Application of high-frequency US to assess vein wall morphology for cannulation readiness. Radiology 2011;261(2):616-62478. Robbin ML, Chamberlain NE, Lockhart ME, et al. Hemodialysis arteriovenous fistula maturity: US evaluation. Radiology 2002:59-6479. Wong V, Ward R, Taylor J, et al. Factors associated with early failure of arteriovenous fistulae for haemodialysis access. European Journal of Vascular & Endovascular Surgery 1996:207-21380. Sottiurai VS, Stephens A, Champagne L, et al. Comparative results of early and delayed cannulation of arteriovenous graft in haemodialysis. European Journal of Vascular and Endovascular Surgery 1997;13(2):139-14181. Feldman L, Shani M, Mursi J, et al. Effect of timing of the first cannulation on survival of arteriovenous hemodialysis grafts. Therapeutic Apheresis & Dialysis: Official Peer-Reviewed Journal of the International Society for Apheresis, the Japanese Society for Apheresis, the Japanese Society for Dialysis Therapy 2013:60-6482. Dawidson IJ, Ar, Rajab A, et al. Early use of the Gore-Tex Stretch Graft. Blood Purification

of 226

1996:337-34483. Glickman MH, Burgess J, Cull D, et al. Prospective multicenter study with a 1-year analysis of a new vascular graft used for early cannulation in patients undergoing hemodialysis. Journal of vascular surgery 2015;62(2):434-44184. Hakaim AG, Scott TE. Durability of early prosthetic dialysis graft cannulation: results of a prospective, nonrandomized clinical trial. Journal of Vascular Surgery 1997:1002-1005; discussion 1005-100685. Fan PY, Schwab SJ. Vascular access: concepts for the 1990s. Journal of the American Society of Nephrology : JASN 1992;3(1):1-1186. National Kidney Foundation. KDOQI clinical practice guidelines and clinical practice recommendations for 2006 updates: Hemodialysis adequacy, peritoneal dialysis adequacy and vascular access. American Journal of Kidney Diseases;2006(48 (Suppl 1)):S1-S32287. Ravani P, Quinn RR, Oliver MJ, et al. (2016) Pre-emptive correction for haemodialysis arteriovenous access stenosis. Cochrane Database of Systematic Reviews, 10.1002/14651858.CD010709.pub2.88. Aragoncillo I, Abad S, Caldes S, et al. (2017) Adding access blood flow surveillance reduces thrombosis and improves arteriovenous fistula patency: a randomized controlled trial. Journal of vascular access, 10.5301/jva.5000700.89. Aragoncillo I, Amezquita Y, Caldes S, et al. (2016) The impact of access blood flow surveillance on reduction of thrombosis in native arteriovenous fistula: a randomized clinical trial. Journal of vascular access, 10.5301/jva.5000461.90. Jindal K, Chan CT, Deziel C, et al. Hemodialysis clinical practice guidelines for the Canadian Society of Nephrology. J Am Soc Nephrol 2006;17(3 Suppl 1):S1-2791. National Institute for Health and Care Excellence. Renal replacement therapy services for adults [Nice Guideline QS72]. https://www.nice.org.uk/guidance/qs72.92. Kumwenda M, Mitra S, Reid C. Clinical practice guideline - Vascular access for haemodialysis 6th Edition. https://renal.org/wp-content/uploads/2017/06/vascular-access.pdf.93. Coleman C, Tuttle L, Teevan C, et al. Antiplatelet agents for the prevention of arteriovenous fistula and graft thrombosis: a meta analysis (Provisional abstract). International Journal of Clinical Practice 2010;64(9):1239-124494. Kaufman JS, Connor TZ, Zhang JH, et al. Randomized controlled trial of clopidogrel plus aspirin to prevent hemodialysis access graft thrombosis. Journal of the American Society of Nephrology 2003;14(9):2313-232195. Kooistra MP, van Es A, Marx JJ, et al. Low-dose aspirin does not prevent thrombovascular accidents in low-risk haemodialysis patients during treatment with recombinant human erythropoietin. Nephrology Dialysis Transplantation 1994;9(8):1115-112096. Kim M, Kim JU, Kim SM, et al. Effectiveness of beraprost sodium in maintaining vascular access patency in patients on hemodialysis. International urology and nephrology 2017;49(7):1287-129597. Sreedhara R, Himmelfarb J, Lazarus J, et al. Anti-platelet therapy in graft thrombosis: results of a prospective, randomized, double-blind study. Kidney International 1994;45(5):1477-148398. Crowther MA, Clase CM, Margetts PJ, et al. Low-intensity warfarin is ineffective for the prevention of PTFE graft failure in patients on hemodialysis: a randomized controlled trial. Journal of the American Society of Nephrology 2002;13(9):2331-233799. Grudzinski A, Mendelssohn D, Pierratos A, et al. A systematic review of buttonhole cannulation practices and outcomes. Seminars in Dialysis 2013;26(4):465-475100. Wong B, Muneer M, Wiebe N, et al. Buttonhole versus rope-ladder cannulation of arteriovenous fistulas for hemodialysis: a systematic review. American Journal of Kidney Diseases 2014;64(6):918-936101. Muir CA, Kotwal SS, Hawley CM, et al. Buttonhole cannulation and clinical outcomes in a home hemodialysis cohort and systematic review. Clinical journal of the American Society of Nephrology 2014;9(1):110-119102. Vaux E, King J, Lloyd S, et al. Effect of buttonhole cannulation with a polycarbonate PEG on in-center hemodialysis fistula outcomes: a randomized controlled trial. American Journal of Kidney Diseases 2013;62(1):81-88

of 226

103. Toma S, Shinzato T, Fukui H, et al. A timesaving method to create a fixed puncture route for the buttonhole technique. Nephrology Dialysis Transplantation 2003;18(10):2118-2121104. Struthers J, Allan A, Peel R, et al. Buttonhole needling of ateriovenous fistulae: a randomized controlled trial. ASAIO journal 2010;56(4):319-322105. Chow J, Rayment G, San Miguel S, et al. A randomised controlled trial of buttonhole cannulation for the prevention of fistula access complications. Journal of Renal Care 2011;37(2):85-93106. MacRae J, Ahmed S, Atkar R, et al. A randomized trial comparing buttonhole with rope ladder needling in conventional hemodialysis patients. Clinical Journal of the American Society of Nephrology 2012;7(10):1632-1638107. MacRae J, Ahmed S, Hemmelgarn B. Arteriovenous fistula survival and needling technique: long-term results from a randomized buttonhole trial. American Journal of Kidney Diseases 2014;63(4):636-642108. Parisotto MT, Schoder VU, Miriunis C, et al. Cannulation technique influences arteriovenous fistula and graft survival. Kidney international 2014;86(4):790-797109. Nesrallah GE, Mustafa RA, MacRae J, et al. Canadian Society of Nephrology guidelines for the management of patients with ESRD treated with intensive hemodialysis. American Journal of Kidney Diseases 2013;62(1):187-198110. Vale E, Lopez-Vargas P, Polkinghorne K. Nursing care of arteriovenous fistula/ arteriovenous graft. http://www.cari.org.au/Dialysis/dialysis%20vascular%20access/Nursing_care_of_AVF_AVG.pdf.111. Upadhyay C, Cameron K, Murphy L, et al. Measuring pain in patients undergoing hemodialysis: a review of pain assessment tools. Clinical Kidney Journal 2014;7(4):367-372112. Moore J, Jarvis E. Comparison of large-gauge hollow-bore haemodialysis access needles: a randomised controlled trial. Nephrology 2015;20(53):60-89113. Morselli C, Chiari P, Aliberti T, et al. Sharp versus blunt dialysis needle use with buttonhole method: Open randomised trial. Journal of Renal Care 2015;41(4):213-221114. Marticorena RM, Dacouris N, Donnelly SM. Randomized pilot study to compare metal needles versus plastic cannulae in the development of complications in hemodialysis access. Journal of Vascular Access 2018;19(3):272-282115. Helthuis J, Haller M, Huisman L, et al. The effect of timing of thrombectomy for occluded autogenous arteriovenous fistulas on functional outcome: A systematic review. Under submission 2018116. Beathard GA. Successful treatment of the chronically thrombosed dialysis access graft: resuscitation of dead grafts. Seminars in Dialysis 2006;19(5):417-420117. Diskin CJ. The importance of timing of surgery for hemodialysis vascular access thrombectomy. Nephron 1997;75(2):233-237118. Sadaghianloo N, Jean-Baptiste E, Gaid H, et al. Early surgical thrombectomy improves salvage of thrombosed vascular accesses. Journal of Vascular Surgery 2014;59(5):1377-1384.e1371-1372119. El-Damanawi R, Kershaw S, Campbell G, et al. Successful restoration of arteriovenous dialysis access patency after late intervention. Clinical Kidney Journal 2015;8(1):82-86120. Graor RA, Risius B, Denny KM, et al. Local thrombolysis in the treatment of thrombosed arteries, bypass grafts, and arteriovenous fistulas. Journal of Vascular Surgery 1985;2(3):406-414121. Beathard GA. Percutaneous transvenous angioplasty in the treatment of vascular access stenosis. Kidney International 1992;42(6):1390-1397122. Cohen MA, Kumpe DA, Durham JD, et al. Improved treatment of thrombosed hemodialysis access sites with thrombolysis and angioplasty. Kidney International 1994;46(5):1375-1380123. Kanterman RY, Vesely TM, Pilgram TK, et al. Dialysis access grafts: anatomic location of venous stenosis and results of angioplasty. Radiology 1995;195(1):135-139124. Trerotola SO, Vesely TM, Lund GB, et al. Treatment of thrombosed hemodialysis access grafts: Arrow-Trerotola percutaneous thrombolytic device versus pulse-spray thrombolysis. Arrow-Trerotola Percutaneous Thrombolytic Device Clinical Trial. Radiology 1998;206(2):403-414125. Maleux G, De Coster B, Laenen A, et al. Percutaneous rheolytic thrombectomy of thrombosed autogenous dialysis fistulas: technical results, clinical outcome, and factors influencing patency. Journal of Endovascular Therapy 2015;22(1):80-86

of 226

126. Marcelin C, D'Souza S, Le Bras Y, et al. Mechanical Thrombectomy in Acute Thrombosis of Dialysis Arteriovenous Fistulae and Grafts Using a Vacuum-Assisted Thrombectomy Catheter: A Multicenter Study. Journal of Vascular and Interventional Radiology 2018;29(7):993-997127. Uflacker R, Rajagopalan P, Selby J, et al. Thrombosed dialysis access grafts: randomized comparison of the Amplatz thrombectomy device and surgical thromboembolectomy. European Radiology 2004;14(11):2009-2014128. Barth KH, Gosnell MR, Palestrant AM, et al. Hydrodynamic thrombectomy system versus pulse-spray thrombolysis for thrombosed hemodialysis grafts: a multicenter prospective randomized comparison. Radiology 2000;217(3):678-684129. Vogel PM, Bansal V, Marshall MW. Thrombosed hemodialysis grafts: lyse and wait with tissue plasminogen activator or urokinase compared to mechanical thrombolysis with the Arrow-Trerotola percutaneous thrombolytic device. Journal of Vascular and Interventional Radiology 2001;12(10):1157-1165130. Lambert G, Freedman J, Jaffe S, et al. Comparison of surgical and radiological interventions for thrombosed arteriovenous access. The journal of vascular access 2018:1129729818762007131. Balshem H, Helfand M, Schünemann HJ, et al. GRADE guidelines: 3. Rating the quality of evidence. Journal of Clinical Epidemiology 2011;64(4):401-406

of 226