A systematic review, meta-analysis, and meta-regression

of the efficacy and safety of endovascular arteriovenous

fistula creation
Ian Jun Yan Wee,a Hao Yun Yap, MBBS, MRCSEd (Surgery), MMed (Surgery), FRCSEd,b
Tjun Yip Tang, MB Bchir, MA (Cantab), FRCS (Glas), FRCS (Eng),b and
Tze Tec Chong, MBBS (Sydney), FACS (General & Vascular Surgery),b Singapore

ABSTRACT
Objective: The percutaneous endovascular approach to arteriovenous fistula (AVF) creation is a minimally invasive
alternative to surgical AVF creation. This systematic review and meta-analysis aimed to investigate the efficacy and safety
of endovascular AVF creation in patients with end-stage renal disease.
Methods: This study conformed to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines.
An electronic search was performed on major databases to identify relevant articles. Meta-analysis of proportions and
meta-regression were conducted.
Results: Seven studies totaling 300 patients were included, of which four evaluated the everlinQ (TVA Medical, Austin,
Tex) and three employed the Ellipsys (Avenu Medical, San Juan Capistrano, Calif) systems. The overall technical success
rate was 97.50% (95% confidence interval [CI], 94.98-99.31%; I2 ¼ 0.00%; P ¼ .487). The 90-day maturation rate was 89.27%
(95% CI, 84.00-93.66%; I2 ¼ 21.29%; P ¼ .283), and the 6-month patency and 12-month patency were 91.99% (95% CI,
87.98-95.35%; I2 ¼ 0.00%; P ¼ .780) and 85.71% (95% CI, 79.90-90.71%; I2 ¼ 0.00%; P ¼ NS), respectively. The overall
procedure-related complication rate was 5.46% (95% CI, 0.310-14.42%; I2 ¼ 81.21%; P ¼ .000). Meta-regression was con-
ducted on the pooled rates of technical success and complication, showing that age, diabetes, white race, hypertension,
on dialysis, and body mass index were not significant sources of heterogeneity.
Conclusions: Current endovascular AVF systems appear to be effective and safe. However, given the lack of head-to-head
comparative analyses with surgical AVF creation, superiority cannot be established. (J Vasc Surg 2020;71:309-17.)
Keywords: Endovascular arteriovenous fistula creation; EverlinQ; Ellipsys

Arteriovenous fistulas (AVFs) are widely regarded as the
most effective access for hemodialysis in patients with
end-stage renal failure. Ever since the introduction of sur-
gical AVF creation by Brescia et al,1 various AVF options
have become widely available. However, these fistulas
are at risk of maturation failure and acute thrombosis.2-5
Multiple interventions may also be necessary to maintain
fistula patency and function.6,7
Highlights of recent technologic advancements include
a minimally invasive, percutaneous endovascular
approach to AVF creation (endoAVF). There are presently
two such devices on the market that have demonstrated
From the Yong Loo Lin School of Medicine, National University of Singaporea;
and the Department of Vascular Surgery, Singapore General Hospital.b
Author conflict of interest: none.
Additional material for this article may be found online at www.jvascsurg.org.
Correspondence: Associate Professor Tze Tec Chong, MBBS (Sydney), FACS
(General & Vascular Surgery), Senior Consultant and Head, Department of
Vascular Surgery, Singapore General Hospital, Outram Rd, Bukit Merah,
Singapore 169608 (e-mail: chong.tze.tec@singhealth.com.sg).
The editors and reviewers of this article have no relevant financial relationships to
disclose per the JVS policy that requires reviewers to decline review of any
manuscript for which they may have a conflict of interest.
0741-5214
Copyright Ó 2019 by the Society for Vascular Surgery. Published by Elsevier Inc.
https://doi.org/10.1016/j.jvs.2019.07.057
encouraging results in clinical trials and cohort
studies.8-14 The Ellipsys Vascular Access System (Avenu
Medical, San Juan Capistrano, Calif) is a thermal resis-
tance anastomosis device that enables arteriovenous
anastomosis of the proximal radial artery and perforating
vein.9,12 The everlinQ endoAVF system (TVA Medical, Aus-
tin, Texas) works slightly differently in that it is a dual
catheter-based system that creates an AVF between
deep arteries (typically the common ulnar artery) and
veins (ulnar vein) of the proximal forearm. Both catheters
contain magnets to facilitate correct alignment between
the artery and vein, and the anastomosis is created using
radiofrequency energy supplied by a generator.11,14
It was hence the aim of this systematic review, meta-
analysis, and meta-regression to evaluate the safety and
efficacy of endoAVF devices for AVF creation in patients
with end-stage renal failure.
METHODS
This study was performed in strict accordance with the
Preferred Reporting Items for Systematic Reviews and
Meta-Analyses guidelines.15
Literature search. A systematic search was performed
on the electronic databases MEDLINE, Embase, and
Cochrane Library from date of inception to February 31,

309
310 Yan Wee et al
2018 to identify all relevant published articles. A combi-
nation of the following search terms using Boolean oper-
ators was used in MEDLINE: (percutaneous.ti,ab. OR
endovascular.ti,ab.) AND (arteriovenous fistula.ti,ab. OR
arteriovenous fistula.mp. OR arteriovenous fistula/). The
full search strategy for all databases can be found in
the Appendix (online only). The reference lists of
included studies were also manually searched to identify
other relevant studies.
Study selection. Two reviewers (I.W., H.Y.) screened the
studies independently for inclusion, and conflicts were
resolved by consensus or by appeal to the senior author
(T.C.). The first stage involved abstract and title screening,
whereas the second required full-text screening to
establish inclusion.
The inclusion criteria included any randomized or non-
randomized study that specifically investigated the
safety and efficacy of endoAVF systems. Studies of the
following design were excluded: non-English studies,
case reports and series (nonconsecutive collection of
data on more than one patient), animal and laboratory
studies, and literature reviews.
Outcomes of interest and data extraction. Using a pro-
forma, two authors (I.W., H.Y.) extracted the following
data independently: baseline characteristics (age, sex,
comorbidities), technical success, 90-day maturation,
cumulative patency at 6 and 12 months, and procedure-
related complications. Fistula failure was defined as
thrombosis, need for surgical intervention, need for
percutaneous intervention, or abandonment of the fistula
for lack of maturation. For purposes of calculating cumu-
lative patency, the duration of patency was defined as the
time from fistula creation until the last follow-up assess-
ment or until fistula failure.16 Technical success was
defined as angiographic evidence of brisk flow within the
AVF and absence of leakage of blood outside the AVF. As
defined by the National Kidney Foundation Kidney Dis-
ease Outcomes Quality Initiative clinical practice guide-
lines and clinical practice recommendations,17 all studies
defined maturation as having the brachial artery flow rate
$500 mL/min and the vein diameter >4 mm. Procedure-
related complication was defined as any unintended
medical occurrence directly arising from the procedure or
device from time of procedure initiation to completion.
Depending on what was reported by the authors in their
respective original papers, this may or may not include
access site complications, pseudoaneurysm, thrombosis,
arterial dissection, closure device embolization, hema-
toma, and steal syndrome.
Risk of bias and quality assessment. The Newcastle-
Ottawa Scale (NOS) was used to assess the quality of
included studies, which can be awarded a maximum of
9 points. Publication bias was not evaluated because of
the small number of studies.
Journal of Vascular Surgery
January 2020
Statistical analyses. For outcomes with more than one
study, a meta-analysis of pooled prevalence (numerator
and denominator of all dichotomous outcomes) and
their 95% confidence intervals (CIs) were obtained on the
basis of the exact binomial distributions with Freeman-
Tukey double arcsine transformation.18 The random-
effects model with inverse variance weighting was used
for all analyses to account for heterogeneity between
studies. Mixed-effects meta-regression was performed to
explore for sources of heterogeneity.19
RESULTS
Seven studies were included with a total of 300
patients, of which four were prospective multicenter
trials and three were retrospective cohort studies8-14
(Fig 1). Four studies evaluated the everlinQ system8,11,13,14;
the remaining three assessed the Ellipsys system.9,10,12
For the everlinQ system, all studies performed an ulnar
artery-ulnar vein anastomosis, and coil embolization of
the entry brachial vein was commonly performed to redi-
rect flow to the superficial veins. In the Ellipsys system,
the anastomosis was commonly made with the proximal
radial artery and the perforating vein. Two studies used
the older 6F system, whereas one study employed the
newer 4F system.8 The overall risk of bias was low
because all studies scored at least 6 of 9 points on the
Newcastle-Ottawa Scale. All studies suffered from
inherent selection and confounding biases, given either
the retrospective or nonrandomized study designs. A
summary of baseline characteristics can be found in
the Table.
Overall efficacy. Meta-analysis of proportions yielded a
technical success rate of 97.50% (95% CI, 94.98-99.31%;
I2 ¼ 0.00%; Fig 2). The 90-day maturation rate was
89.27% (95% CI, 84.00-93.66%; I2 ¼ 21.29%; Fig 3). In terms
of patency, the 6-month patency and 12-month patency
were 91.99% (95% CI, 87.98-95.35%; I2 ¼ 0.00%; Fig 4) and
85.71% (95% CI, 79.90-90.71%; I2 ¼ .00%; Fig 5), respec-
tively. The overall procedure-related complication rate
was 5.46% (95% CI, 0.310-14.42%; I2 ¼ 81.21%; Fig 6).
everlinQ system. The meta-analysis of pooled technical
success rate was 99.45% (95% CI, 96.46-100.00%; I2 ¼
0.00%; Supplementary Fig 1, online only). The pooled
90-day maturation rate and 6-month patency were
88.17% (95% CI, 80.51-94.24%; I2 ¼ 0.00%; Supplementary
Fig 2, online only) and 92.61% (95% CI, 86.47-97.26%; I2 ¼
0.00%; Supplementary Fig 3, online only), respectively.
The rate of procedure-related complication was 8.59%
(95% CI, 2.96-16.15%; I2 ¼ 28.76%; Supplementary Fig 4,
online only).
Ellipsys system. The pooled technical success rate was
95.19% (95% CI, 91.07-98.23%; I2 ¼ 0.00%; Supplementary
Fig 5, online only). The pooled 90-day maturation rate
and 6-month patency were 89.35% (95% CI, 83.53-94.11%;
Journal of Vascular Surgery
Volume 71, Number 1
Fig 1. Preferred Reporting Items for Systematic Reviews
and Meta-Analyses flow diagram.
I2 ¼ 0.00%; Supplementary Fig 6, online only) and
90.98% (95% CI, 85.45-95.38%; I2 ¼ 0.00%;
Supplementary Fig 7, online only), respectively. The
rate of procedure-related complication was 2.48% (95%
CI, 0.00-16.23%; I2 ¼ 86.85%; Supplementary Fig 8,
online only).
Meta-regression for overall efficacy. Meta-regression
was conducted on the pooled rate of technical success,
which showed that the following patient characteristics
were not significant sources of heterogeneity: age
(b ¼ 1.099), male sex (b ¼ 1.033), white race (b ¼ .935),
Yan Wee et al 311
diabetes (b ¼ .904), previously dialyzed on catheters
(b ¼ .970), hypertension (b ¼ 1.019), and body mass index
(b ¼ 1.147). In view of heterogeneity in the pooled rates of
complication, a meta-regression was also performed,
demonstrating no significant sources of heterogeneity
for the following variables: age (b ¼ .937), male sex (b ¼
.972), diabetes (b ¼ 1.061), hypertension (b ¼ 1.000), and
body mass index (b ¼ .669).
DISCUSSION
To our knowledge, this is the most up to date system-
atic review and meta-analysis investigating the efficacy
and safety of percutaneous endoAVF systems. The study
suggests that the short-term rates of technical success,
maturation, and patency are high, with a low risk of
procedure-related complications.
Both the everlinQ and Ellipsys systems boast unique
capabilities. For the everlinQ system, the earlier Novel
Endovascular Access Trial (NEAT) and FLEX trial used
the 6F first-generation device, but a rapid-exchange
4F 0.014-inch option was recently made available.
With the newer system, visual markers and magnets
are now optimized, coupled with a shorter radiofre-
quency energy release time (0.7 second compared
with 2 seconds).8 Furthermore, the smaller caliber
design allows access points distally at the wrist. An
“antiparallel approach” can also be employed, whereby
access is gained through a wrist artery while the
brachial vein is accessed from above, capitalizing on
the system’s bidirectional flow to facilitate arterial
hemostasis. The procedure requires only angiography
of the antecubital fossa and forearm at the start of
the procedure, hence reducing the contrast material
volume required.11,13,14
The Ellipsys system involves retrograde access into
the cubital or brachial vein using a single 6F tissue
fusion catheter, designed for a side-to-side arteriove-
nous anastomosis facilitated by the simultaneous
induction of pressure and heat. Furthermore, as the
procedure is carried out under ultrasound guidance,
it avoids the need for contrast material and radiation
exposure.10,12
The high maturation rate in this study could be
attributed to the minimally invasive nature of endoAVF
techniques, which reduces surgical trauma by avoiding
skin and soft tissue incision, vessel transposition and
manipulation, side branch ligation, and suturing. These
are known contributory risk factors to the development
of neointimal hyperplasia, the Achilles heel of fistula
maturation.20,21 Nevertheless, large cohort studies from
the United States22 and The Netherlands23 have reported
maturation rates ranging from 64% to 76% in surgical
AVFs. In our experience published last year, we
312 Yan Wee et al Journal of Vascular Surgery
January 2020

Table. Baseline characteristic, comorbidities, and perioperative parameters

First author, year Study design Device No. of patients Age, years Male, % White, % DM, % Dialyzed, %
Rajan,14
2015 (FLEX) RC everlinQ 33 51.0 6 11.4 61.0 NR 58.0 NR
Lok,11 2017 (NEAT) PT everlinQ 60 59.9 6 13.6 65.0 60.0 50.0 43.0

Radosa,13 2017 RC everlinQ 8 57.0 (25.0-72.0) 75.0 NR 50.0 NR

Berland,8 2019 (EASE)a PT everlinQ 32 51.4 6 12.7 97.0 NR 56.3 3.0

Hull,9 2017 PT Ellipsys 26 45.5 6 13.6 38.4 NR 65.0 100

Hull,10 2018 PT Ellipsys 107 56.7 6 12.0 72.9 73.8 64.5 61.7

Mallios,12 2018 RC Ellipsys 34 64.0 (22.0-89.0) 66.7 75.0 63.0 69.0

AV, Arteriovenous; BMI, body mass index; DM, diabetes mellitus; HTN, hypertension; NR, not reported; PT, prospective trial; RC, retrospective cohort.
Categorical variables are presented as number (%). Continuous variables are presented as mean 6 standard deviation or median (interquartile
range).
a
Conference abstract.

demonstrated a maturation rate of 78%.24 However,
given the lack of comparative data and the small sample
size in this meta-analysis, caution must be taken to justify
its superiority over surgical AVFs.
The other important outcome that was not meta-
analyzed because of the lack of data is the rate of sec-
ondary procedures. In one everlinQ study by Lok et al,11
the rate of secondary procedure was 0.46 per patient
per year. Common procedures included basilic vein
embolization, transposition, and embolization of a trib-
utary vein. The only other study was conducted by
Hull et al,10 showing that the mean number of proced-
ures per patient per year was two with the Ellipsys
system. Despite the relatively lower rates than those
reported in the surgical AVF literature,5 comparative
studies are again still necessary to establish these
findings.
As such, under what circumstances should the
endoAVF approach be regarded as first line? Unless
the traditional radiocephalic fistula option is unavai-
lable, the available evidence does not support this as
the first-line approach. Patients should also have a min-
imal access target vessel of 2 mm in diameter and a
perforator vein of >3 mm. Furthermore, the proximity
between the artery and adjacent vein should be
>1.5 mm (Table). However, the endoAVF approach may
Journal of Vascular Surgery Yan Wee et al 313
Volume 71, Number 1

Table. Continued.
Intraoperative adjunctive
HTN, % BMI, kg/m2 AV anastomosis Preoperative requirement procedures
NR 24.3 6 3.8 Ulnar artery-vein NR NR
92.0 27.9 6 6.1 Ulnar artery-vein Routine vessel mapping Coil embolization of entry
brachial vein
75.0 24.5 (22.8-30.2) Ulnar artery-vein Brachial artery and vein Coil embolization of entry
$2.5 mm brachial vein
Proximal ulnar artery and vein
$2.0 mm
Proximal ulnar artery and vein
have to run in parallel and in
proximity $2 cm
Perforating vein
Cephalic or basilica vein
$2.0 mm
90.0 >25: 51.6% Ulnar artery-vein (37.5%) NR NR
Radial artery-vein (62.5%)
92.0 26.7 6 5.1 Proximal radial Radial artery >2.0 mm Coil embolization of entry
artery-perforating vein brachial vein
Adjacent vein >2.0 mm Banding of median basilic vein
Adequate collateral arterial
perfusion
Radial artery-adjacent vein
proximity >1.5 mm
98.1 31.2 6 NR Proximal radial Radial artery >2.0 mm Coil embolization of entry
(18.3-48.9) artery-perforating vein brachial vein
Adjacent vein >2.0 mm Banding of median basilic vein
Adequate collateral arterial
perfusion
Radial artery-adjacent vein
proximity >1.5 mm
NR >30: 39.0% Proximal radial Radial artery >2.0 mm Percutaneous transluminal
artery-perforating vein angioplasty
Perforating vein >3.0 mm Coil embolization
Radial artery-adjacent vein
proximity >1.5 mm

not be necessarily contraindicated in patients with
previously failed surgical AVFs because endoAVF is per-
formed more distally at the wrist, whereas the surgical
AVF was likely to have been created proximally above
the elbow.
The other paramount consideration is the difference
in cannulating these deeper endoAVFs, which could
pose a potential challenge for inexperienced dialysis
centers and nurses. Strategies have been described
by various authors to overcome this shortcoming.
Mallios et al,12 for instance, recommended
ultrasound-guided cannulation. Other more invasive
adjunctive procedures were reported by Hull et al10 to
redirect flow from deep to superficial veins, hence
bringing the matured veins closer to the skin surface
for cannulation. These included brachial vein emboli-
zation or ligation, valvulotomy, and transposition.
Regardless, it is necessary for dialysis centers and
nurses to be educated on the cannulation process as
they play a crucial role in ensuring proper technique
and longevity of the AVF.25
The issue of cost has not yet been addressed, although
it theoretically can be reduced with the endoAVF
systems.26 They can be done as day-case procedures,
even in outpatient settings, without the need for general
anesthesia and an operating room.
314 Yan Wee et al Journal of Vascular Surgery
January 2020

Fig 2. Pooled technical success rates. CI, Confidence interval; ES, effect size.

Fig 3. Pooled maturation rates. CI, Confidence interval; ES, effect size.

The findings of this study must be interpreted in the
context of known limitations. As all included studies
were nonrandomized, they are subject to inherent selec-
tion, confounding, and residual biases. For instance, one
study had a disproportionately low proportion of blacks,11
whereas one study included a population of patients
with a lower mean body mass index14 compared with
the general population in the United States. Next, there
is a lack of long-term data because of the short follow-
up in some studies.9,13,14 Furthermore, as all studies had
Journal of Vascular Surgery Yan Wee et al 315
Volume 71, Number 1

Fig 4. Pooled 6-month patency rates. CI, Confidence interval; ES, effect size.

Fig 5. Pooled 12-month patency rates. CI, Confidence interval; ES, effect size.

only a single arm, comparative meta-analyses were not CONCLUSIONS

feasible. Last, other key outcomes were not meta-
analyzed in this study because of the lack of data. For
instance, the measurement of fistula function by its abil-
ity to be used for dialysis was not included as an
outcome and should be reported in future trials.
Given the encouraging efficacy and safety of
current endoAVF systems, it could be an alternative
for patients with the right indications but presently
should not be regarded as first line, given the
lack of comparative, high-quality evidence. Much
316 Yan Wee et al
Fig 6. Pooled procedure-related complication rates. CI, Confidence interval; ES, effect size.
awaits to be seen with prospective head-to-head
trials and the continuous development of such
technology.
AUTHOR CONTRIBUTIONS
Conception and design: IW, HY, TC
Analysis and interpretation: IW, HY, TY, TC
Data collection: IW, HY, TY, TC
Writing the article: IW, HY, TC
Critical revision of the article: IW, HY, TY, TC
Final approval of the article: IW, HY, TY, TC
Statistical analysis: IW, HY
Obtained funding: Not applicable
Overall responsibility: TC
IW and HY contributed equally to this article and share
co-first authorship
REFERENCES
1. Brescia MJ, Cimino JE, Appell K, Hurwich BJ, Scribner BH.
Chronic hemodialysis using venipuncture and a surgically
created arteriovenous fistula. 1966. J Am Soc Nephrol
1999;10:193-9.
2. Al-Jaishi AA, Oliver MJ, Thomas SM, Lok CE, Zhang JC,
Garg AX, et al. Patency rates of the arteriovenous fistula for
hemodialysis: a systematic review and meta-analysis. Am J
Kidney Dis 2014;63:464-78.
3. Peterson WJ, Barker J, Allon M. Disparities in fistula matu-
ration persist despite preoperative vascular mapping. Clin J
Am Soc Nephrol 2008;3:437-41.
4. Dember LM, Beck GJ, Allon M, Delmez JA, Dixon BS,
Greenberg A, et al. Effect of clopidogrel on early failure of
arteriovenous fistulas for hemodialysis: a randomized
controlled trial. JAMA 2008;299:2164-71.
Journal of Vascular Surgery
January 2020
5. Falk A. Maintenance and salvage of arteriovenous fistulas.
J Vasc Interv Radiol 2006;17:807-13.
6. Lee T, Barker J, Allon M. Tunneled catheters in hemodialysis
patients: reasons and subsequent outcomes. Am J Kidney
Dis 2005;46:501-8.
7. Biuckians A, Scott EC, Meier GH, Panneton JM,
Glickman MH. The natural history of autologous fistulas as
first-time dialysis access in the KDOQI era. J Vasc Surg
2008;47:415-21; discussion: 420-1.
8. Berland T, Westin G, Clement J, Griffin J, Sadek M,
Blumberg S, et al. Endovascular creation of an arteriovenous
fistula with a next generation 4 Fr device design for hemo-
dialysis access: clinical experience from the EASE study. Ann
Vasc Surg 2019;55:7.
9. Hull JE, Elizondo-Riojas G, Bishop W, Voneida-Reyna YL.
Thermal resistance anastomosis device for the percuta-
neous creation of arteriovenous fistulae for hemodialysis.
J Vasc Interv Radiol 2017;28:380-7.
10. Hull JE, Jennings WC, Cooper RI, Waheed U,
Schaefer ME, Narayan R. The pivotal multicenter trial of
ultrasound-guided percutaneous arteriovenous fistula
creation for hemodialysis access. J Vasc Interv Radiol
2018;29:149-58.e5.
11. Lok C, Rajan DK, Buldo G, Rajan D, Sniderman K, Jaskolka J,
et al. Endovascular proximal forearm arteriovenous fistula
for hemodialysis access: results of the prospective, multi-
center Novel Endovascular Access Trial (NEAT). Am J Kidney
Dis 2017;70:486-97.
12. Mallios A, Jennings WC, Boura B, Costanzo A, Bourquelot P,
Combes M. Early results of percutaneous arteriovenous fis-
tula creation with the Ellipsys Vascular Access System.
J Vasc Surg 2018;68:1150-6.
13. Radosa CG, Radosa JC, Weiss N, Schmidt C, Werth S,
Hofmockel T, et al. Endovascular creation of an arteriove-
nous fistula (endoAVF) for hemodialysis access: first results.
Cardiovasc Intervent Radiol 2017;40:1545-51.
Journal of Vascular Surgery
Volume 71, Number 1
14. Rajan DK, Ebner A, Desai SB, Rios JM, Cohn WE. Percuta-
neous creation of an arteriovenous fistula for hemodialysis
access. J Vasc Interv Radiol 2015;26:484-90.
15. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred report-
ing items for systematic reviews and meta-analyses: the
PRISMA statement. PLoS Med 2009;6:e1000097.
16. Lee T, Mokrzycki M, Moist L, Maya I, Vazquez M, Lok CE.
Standardized definitions for hemodialysis vascular access.
Semin Dial 2011;24:515-24.
17. National Kidney Foundation. KDOQI clinical practice
guidelines and clinical practice recommendations for
2006 updates: hemodialysis adequacy, peritoneal dialysis
adequacy and vascular access. Am J Kidney Dis 2006;48:
1-322.
18. Nyaga VN, Arbyn M, Aerts M. Metaprop: a Stata command to
perform meta-analysis of binomial data. Arch Public Health
2014;72:39.
19. Harbord RM, Higgins JPT. Meta-regression in Stata. Stata J
2008;8:493-519.
20. Roy-Chaudhury P, Arend L, Zhang J, Krishnamoorthy M,
Wang Y, Banerjee R, et al. Neointimal hyperplasia in early
arteriovenous fistula failure. Am J Kidney Dis 2007;50:
782-90.
21. Roy-Chaudhury P, Sukhatme VP, Cheung AK. Hemodialysis
vascular access dysfunction: a cellular and molecular view-
point. J Am Soc Nephrol 2006;17:1112-27.
Yan Wee et al 317
22. Woodside KJ, Bell S, Mukhopadhyay P, Repeck KJ,
Robinson IT, Eckard AR, et al. Arteriovenous fistula matura-
tion in prevalent hemodialysis patients in the United States:
a national study. Am J Kidney Dis 2018;71:793-801.
23. Voorzaat BM, van der Bogt KEA, Janmaat CJ, van Schaik J,
Dekker FW, Rotmans JI, et al. Arteriovenous fistula matura-
tion failure in a large cohort of hemodialysis patients in the
Netherlands. World J Surg 2018;42:1895-903.
24. Yap HY, Pang SC, Tan CS, Tan YL, Goh N, Achudan S, et al.
Catheter-related complications and survival among incident
hemodialysis patients in Singapore. J Vasc Access 2018;19:
602-8.
25. Parisotto MT, Pelliccia F, Grassmann A, Marcelli D. Elements
of dialysis nursing practice associated with successful
cannulation: result of an international survey. J Vasc Access
2017;18:114-9.
26. Yang S, Lok C, Arnold R, Rajan D, Glickman M. Comparison of
post-creation procedures and costs between surgical and
an endovascular approach to arteriovenous fistula creation.
J Vasc Access 2017;18(Suppl 2):8-14.
Submitted Feb 25, 2019; accepted Jul 15, 2019.
Additional material for this article may be found online
at www.jvascsurg.org.
317.e1 Yan Wee et al Journal of Vascular Surgery
January 2020

Appendix (online only). Search strategy.

Journal of Vascular Surgery Yan Wee et al 317.e2
Volume 71, Number 1

Supplementary Fig 1 (online only). Pooled technical success rates for everlinQ system. CI, Confidence interval; ES,
effect size.

Supplementary Fig 2 (online only). Pooled maturation rates for everlinQ system. CI, Confidence interval; ES,
effect size.
317.e3 Yan Wee et al Journal of Vascular Surgery
January 2020

Supplementary Fig 3 (online only). Pooled 6-month patency rates for everlinQ system. CI, Confidence interval;
ES, effect size.

Supplementary Fig 4 (online only). Pooled procedure-related complication rates for everlinQ system. CI, Con-
fidence interval; ES, effect size.
Journal of Vascular Surgery Yan Wee et al 317.e4
Volume 71, Number 1

Supplementary Fig 5 (online only). Pooled technical success rates for Ellipsys system. CI, Confidence interval; ES,
effect size.

Supplementary Fig 6 (online only). Pooled maturation rates for Ellipsys system. CI, Confidence interval; ES, effect
size.
317.e5 Yan Wee et al Journal of Vascular Surgery
January 2020

Supplementary Fig 7 (online only). Pooled 6-month patency rates for Ellipsys system. CI, Confidence interval; ES,
effect size.

Supplementary Fig 8 (online only). Pooled procedure-related complication rates for Ellipsys system. CI, Confi-
dence interval; ES, effect size.