SPECIAL ARTICLE

Hemodialysis Vascular Access Training and Practices Are Key to

Improved Access Outcomes
David A. Goodkin, MD,1 Ronald L. Pisoni, PhD, MS,1 Francesco Locatelli, MD,2
Friedrich K. Port, MD, MS,1 and Rajiv Saran, MD, MS3
Recognizing that autologous arteriovenous fistula use was associated with improved outcomes in
hemodialysis patients, the 1997 Dialysis Outcomes Quality Initiative (DOQI) vascular access practice
guidelines from the National Kidney Foundation stressed fistulas as the optimal means of dialysis
vascular access. In the United States, this emphasis has continued with the Fistula First Breakthrough
Initiative. Much of the data supporting fistulas for dialysis access are derived from longitudinal cohorts,
including the Dialysis Outcomes and Practice Patterns Study (DOPPS), dialysis provider databases,
and other sources. This article reviews major findings from these data sources, focusing on specific
practices and characteristics associated with greater arteriovenous fistula use in dialysis facilities
worldwide. Important and often overlooked characteristics that are discussed in detail include specific
preferences of dialysis staff regarding access type and the emphasis placed on fistula primacy and the
number of fistulas created during surgical training. For example, in the DOPPS, the risk of initial fistula
failure was 34% lower when fistulas were placed by surgeons who had created at least 25 fistulas during
training (P 0.002). It is imperative that dialysis clinicians advocate actively for specific dialysis access
types on behalf of individual patients. Vascular surgery teaching programs must supervise adequate
numbers of fistula procedures for every trainee.
Am J Kidney Dis 56:1032-1042. 2010 by the National Kidney Foundation, Inc.
INDEX WORDS: Vascular access; hemodialysis; arteriovenous fistula; surgical training; mortality.

n 1997, the National Kidney Foundations original Dialysis Outcomes Quality Initiative
(DOQI) vascular access practice guidelines sought
to increase the placement of autologous arteriovenous fistulas in US hemodialysis (HD) patients.1
The DOQI set goals of at least 50% fistula rates for
new (incident) patients and 40% for prevalent patients. In the United States in 2003, the Centers for
Medicare & Medicaid Services, the End-Stage Renal Disease (ESRD) Networks, and key provider
representatives jointly recommended adoption of a
National Vascular Access Improvement Initiative
to achieve the DOQI goals. In 2005, this effort
From the 1Arbor Research Collaborative for Health, Ann
Arbor, MI; 2Department of Nephrology, Dialysis and Transplantation, Alessandro Manzoni Hospital, Lecco, Italy; and
3
Division of Nephrology, Department of Internal Medicine
and Kidney Epidemiology and Cost Center, University of
Michigan, Ann Arbor, MI.
Received January 22, 2010. Accepted in revised form
August 2, 2010. Originally published online as doi:10.1053/
j.ajkd.2010.08.010 on October 21, 2010.
Address correspondence to David A. Goodkin, MD, 3807
134th Ave NE, Bellevue, WA 98005. E-mail: davidagoodkin@
comcast.net
2010 by the National Kidney Foundation, Inc.
0272-6386/10/5606-0005$36.00/0
doi:10.1053/j.ajkd.2010.08.010
1032

transitioned into the Fistula First Breakthrough

Initiative (FFBI), with delineated goals that included increasing the fistula rate in prevalent patients to 66% and decreasing both catheter use and
vascular access complications in HD patients.2
There has been some measure of success to date,
with the FFBI reporting an increase in US fistula
prevalence to 55% in March 2010.3 To reach this
goal, members of the dialysis community are striving to understand the barriers to more widespread
use of fistulas, which could reflect any combination
of patient, provider, and system factors.
Given widely differing fistula rates worldwide, it can be helpful to examine practices
across health care cultures and systems when
attempting to discern optimal vascular access
strategies. In addition to discussing local data
sources and guideline statements, the present
review uses data generated from the Dialysis
Outcomes and Practice Patterns Study (DOPPS),
a large prospective observational study of HD
patients in 12 nations that collects detailed aspects of demographics, comorbid conditions, laboratory values, prescriptions, and dialysis practices at both patient and facility levels.4,5 This
article reviews vascular access practice differences across nations; considers factors associ-

American Journal of Kidney Diseases, Vol 56, No 6 (December), 2010: pp 1032-1042

Vascular Access: Call to Arms

ated with varying use of the 3 major access

types, namely arteriovenous fistula, arteriovenous graft (AVG), and catheter; briefly compares the outcomes (including thrombosis, hospitalization, and mortality) associated with each
access type; and suggests clinical approaches
that could enhance fistula creation, highlighting
strategies that have been successful in health
care systems worldwide. Overall, current data
strongly suggest that intensified demand for fistulas by dialysis clinicians and meticulous vascular
surgery performed by thoroughly trained operators can result in superior patient outcomes, and
these findings are emphasized.

MORTALITY AND MORBIDITY: THE

INFLUENCE OF VASCULAR ACCESS
In multiple observational studies, the risk of
mortality associated with the use of catheters and
AVGs for vascular access universally exceeds that
seen with fistulas.6-9 For example, in the DOPPS,
the relative risk of death was 15% higher for HD
patients with an AVG than for those with a fistula
(P 0.001) in more than 28,000 patients from
more than 300 dialysis units worldwide.6 Individuals receiving maintenance dialysis with a percutaneous catheter fared even worse, with a 32% higher
relative risk of death compared with the fistula
group (P 0.0001) after adjustment for case mix,

1033

comorbid conditions, and laboratory values. Despite extensive adjustment, it is possible that unmeasured confounding factors influenced these mortality differences, with sicker patients a priori both
unable to undergo creation of a fistula and more
likely to die. Although patients in this study dialyzing with a catheter or an AVG were older and had
higher prevalences of numerous comorbid conditions compared with patients dialyzing with a fistula, examining case-mixadjusted facility vascular
access percentages using instrumental variable analysis showed similar results. These findings are
corroborated by studies from the US Renal Data
System (USRDS),7 Australian and New Zealand
Dialysis and Transplant Association Registry,8 and
ESRD Network 6 in the United States.9
Differences in mortality have long been recognized in HD patients in Japan, Europe, and the
United States,10 and differences in patient demographics and burdens of comorbid disease have
explained only a portion of the varying risks of
death,11 suggesting that differing practice patterns also may account for variability in mortality. In Fig 1, mortality risk is compared between
Europe and the United States, first unadjusted,
then adjusted for case mix and select laboratory
values, and last, also adjusted for access type.6
As indicated in Fig 1, there would be no significant difference in mortality risk between Euro-

Figure 1. Case-mixadjusted mortality hazard ratio (HR) for hemodialysis (HD) patients in the United States versus Europe
(EUR), with and without adjustment for differences in facility vascular access use. The HR of mortality for HD patients in the United
States versus EUR (n 24,398) stratified by study phase is shown after different levels of adjustment: unadjusted; adjusted for
patient age, sex, black race, number of years with end-stage renal disease, body weight, 14 summary comorbid conditions,
whether treated in a hospital-based unit, facility median treatment time, facility percentage of patients with serum phosphorus level
5.5 mg/dL, and facility percentage of patients with serum calcium level 10 mg/dL; and further adjusted for percentage of facility
vascular access use plus the previous 23 adjustments. All models accounted for facility clustering effects. EUR refers to France,
Germany, Italy, Spain, and the United Kingdom. Data source: Pisoni et al.6

1034

pean and US patients matched for clinical characteristics if the pattern of vascular access use for
US patients was similar to the pattern in Europe
(ie, if fistula use was increased and catheter and
AVG use was decreased in the United States).
Vascular access practice differences also accounted for almost 30% of the greater US mortality compared with Japan. In a recent study from
Fresenius units in the United States, researchers
showed that catheters are associated with the
greatest mortality risk; fistulas, with the lowest
mortality; and importantly, changing from a catheter to a fistula or AVG, with significantly improved survival.12 A recent analysis from the
DOPPS similarly showed a survival benefit associated with conversion from a catheter to a permanent access in incident HD patients.13 Clearly,
these analyses reinforce the impetus to maximize
the use of fistulas, decrease the use of catheters,
and provide support for the efforts of the FFBI
and other renal advocacy groups.
Mortality is not the only adverse consequence
associated with using AVGs or catheters rather
than fistulas. Fistula use also is associated with
lower morbidity. Risks of all-cause hospitalization, hospitalization for any infection, or hospitalization for vascular accessrelated infection (accounting for 18.6% of all infection-related
hospitalizations) were substantially greater with
increased facility catheter or AVG use compared
with fistula use.6 The risk of vascular access
related infection was 5- to 7-fold higher for
catheters than for fistulas.14 Vascular access survival also is superior with fistulas compared with
AVGs.15-18 The aggregate costs of access repair
and replacement are extremely high. According
to the USRDS, in 2007, dialysis access event
costs per patient per year were $3,194 for patients with fistulas, $5,960 for patients with catheters, and $7,451 for patients with AVGs.18 The
burden on HD patients related to access complications, including pain, anxiety, hospitalization,
and sepsis, are considerable. Again, the mandate
to maximize the use of fistulas and minimize the
use of catheters is compelling.

VARIABILITY IN VASCULAR ACCESS

PLACEMENT: ROOM FOR IMPROVEMENT
There also is strong evidence that improvements in access outcomes are feasible. The FFBI

Goodkin et al

reports that in prevalent US HD patients, fistula

use increased from 32% in July 2003 to 55% in
March 2010, whereas AVG use decreased from
40% to 21% and catheter use decreased from
27% to 24%.3 In addition, multinational DOPPS
data have shown wide variation in the distribution of HD access types used, both within and
between regions, and shed light on practices
associated with superior or inferior access outcomes. Figure 2 shows the percentage of patients
dialyzing through a fistula across facilities.6
Within every geographic region, the prevalence
of each type of access varied tremendously across
facilities. In 2003, mean facility use of fistulas in
Japan was 92%, whereas in North America, it
was only 36%. Remarkably, a recent study found
that 60% of newly created fistulas in 877 patients
at 9 US centers proved unsuitable for HD.19 This
contrasts markedly with results reported for 5,007
new fistulas at 23 Japanese facilities: primary
failure occurred in only 7.6% of fistulas, and
70% of these were salvaged.20 Mean facility use
of catheters in Japan was only 2%, whereas in
North America, on average, 28% of patients
were dialyzing using catheters.6 Factors associated with increased prevalence of fistula use in
the DOPPS include younger age, male sex, lower
body mass index, nondiabetic status, lack of
peripheral vascular disease, and lack of angina.15
Patients in the United States are older and more
commonly have comorbid conditions that decrease the likelihood of fistula creation, but even
after adjustment for these imbalances, the likelihood of having a fistula as opposed to an AVG
was strikingly higher in Europe than in the United
States (adjusted odds ratio [AOR], 21; P
0.0001).15
As shown in Fig 3, the prevalence of fistulas
for hemoaccess in US HD patients in the DOPPS
has increased from 24% in 1996-2000 to 31% in
2002-2003 and 45% in 2005-2007, with AVG
prevalence decreasing from 58% to 41% to 28%
during the same periods.21 Thus, despite having
older patients who bear a greater burden of
comorbid conditions, such as diabetes mellitus
and peripheral vascular disease, compared with
other nations,11 clinicians are capable of increasing the fistula rate in the United States. Unfortunately, catheter use also increased from 17% to
27% during this time frame. In addition, it is
disconcerting that several nations participating

Vascular Access: Call to Arms

1035

Figure 2. Distribution of percentages of facility patients using a fistula by region in Dialysis Outcomes and Practice
Patterns Study II (DOPPS II). Percentages of facility fistula use were determined from access use in a prevalent
cross-section of hemodialysis (HD) patients within each facility at entry. Analyses were restricted to 312 facilities reporting
vascular access use at study entry for a minimum of 13 patients (median, 27 patients, with 20-40 patients in 90% of
facilities). Values in inset are median (25th, 75th percentile). Abbreviations: AVF, arteriovenous fistula; North Am, Canada
and United States; EUR/ANZ, Belgium, France, Germany, Italy, Spain, Sweden, United Kingdom, Australia, and New
Zealand. Adapted from Pisoni et al6 with permission of the National Kidney Foundation.

in the DOPPS with historically excellent vascular access results, such as Italy, Germany, and
Spain, have shown decreasing prevalences of
fistula use and increasing catheter use during the
same time frame (Italian data are shown in Fig
3). Even in Japan, the prevalence of fistula use
decreased from 93% to 90% and use of AVGs
increased from 3% to 7%, although catheter use
remained the lowest of any country at 2%. These
findings of variability in access type across facilities/nations and increased fistula placement during recent years in the United States suggest that

Figure 3. Trends in vascular access use in the United

States and Italy: Dialysis Outcomes and Practice Patterns
Study (DOPPS). Vascular access use is shown for the United
States and Italy based on a
prevalent cross-section of patients participating in DOPPS
I (1996-2000), DOPPS II
(2002-2004), and DOPPS III
(2005-2007). Abbreviations:
AVF, arteriovenous fistula;
AVG, arteriovenous graft.
Data source: Ethier et al.21

there is substantial room for improvement in

many dialysis units.

A CALL TO (FISTULA) ARMS: MODIFIABLE

PRACTICES TO IMPROVE FISTULA CREATION
Advocacy by Nephrologists and Dialysis Nurses
Alters Outcomes
Although clinicians cannot change many patient factors that affect the odds of successful
fistula placement, such as age, sex, vessel size, or
disease history (diabetes mellitus, peripheral vas-

1036

cular disease, etc), there are modifiable factors

that can be addressed that may explain some of
the variability in worldwide fistula rates.15 Most
notable was the striking finding in the first publication from the DOPPS that dialysis facility staff
preference for fistula or AVG was the single
factor associated most strongly with the likelihood of US DOPPS patients dialyzing using a
fistula versus an AVG.22 Remarkably, in 19961999, a total of 41% of US dialysis facility nurse
managers and 22% of physician medical directors did not prefer fistulas as the permanent
access for patients starting HD therapy. Not
unexpectedly, units in which AVGs were preferred had a significantly higher prevalence of
AVG use.23 That facilities that stated a preference for fistulas succeeded in attaining them
more often is a strong signal that there are
modifiable practices. By 2003, sentiment had
swayed: only 7% of US DOPPS nurse managers
still indicated a preference for AVGs,24 and as
described, the prevalence of fistulas steadily increased. This may reflect ongoing educational
efforts on both the local and national levels in the
United States.25 As evidenced by these patterns,
dialysis clinicians must be watchdogs for our
patients and lobby for fistula placement whenever it is feasible, even if this means referring
patients outside the local medical center to a
surgeon with the proven ability to create durable
fistulas.
Rigorous Surgical Training and Adequate
Experience in Fistula Creation Are Critical
The role of surgeon training and exposure to
fistula creation is explored best in a setting in
which multiple strategies are prevalent and international comparisons can occur. Accordingly, the
DOPPS collected comprehensive data relating
surgical expertise to vascular access outcomes.26
These data show a critical role for training, such
that fistula patency is significantly greater for
fistulas placed by surgeons who created at least
25 fistulas during their surgical training (Fig 4).
The risk of primary fistula failure was 34% lower
when created by surgeons who exceeded this
threshold during training (relative risk, 0.66; P
0.002). Similarly, increased surgeon training in
fistula placement was associated with a greater
likelihood of fistula versus AVG placement in
DOPPS patients, with an AOR of 2.2 for fistula

Goodkin et al

Figure 4. Time to primary fistula (arteriovenous fistula

[AVF]) failure in hemodialysis patients for tertiles of the
number of AVFs created by the facilitys primary surgeon
during surgical training. Model accounted for facility clustering effects and was adjusted for age, sex, black (vs other)
race, 14 comorbid conditions, body mass index, time with
end-stage renal disease, prior catheter use, and country.
Primary failure reflects the first intervention to salvage an
access. The relative risk of fistula failure for those placed
by surgeons who had created at least 25 fistulas during
training versus those placed by surgeons who had created
fewer than 25 fistulas during training was 0.66 (P 0.002).
Adapted from Saran et al26 with permission of Wolters
Kluwer Health.

placement for each 2-fold-higher number of fistulas created during training (P 0.0001). In
DOPPS data, surgeons in the United States receive the least vascular access training, with the
average trainee placing only 50 HD accesses, of
which only 16 are fistulas (Fig 5A). Therefore, it
is not surprising that the prevalence of fistulas in
the United States is the lowest and the prevalence
of catheters is the highest of any of the 12 nations
in the DOPPS. Other studies also have examined
the relation between the surgeons who create
fistulas and outcomes. A study of the patency of
108 fistulas created by 1 of 7 surgeons found in
multivariate regression analysis adjusted for patient demographics and diabetes that only the
specific surgeon was a continuous significant
determinant of fistula patency.27 A subsequent
report also concluded that surgeon selection had
a significant impact on fistula placement and
survival, although this study compared only 2
vascular access surgeons and considered only 75
patients.28
Concerns exist regarding successful access
placement by surgical trainees. Early in the
DOPPS experience, 1 study showed that fistula
use was substantially lower than AVG use when

Vascular Access: Call to Arms

1037

600
56

Catheter

79

AVG

# VA Created

500

AVF

400
300

35
426

200
24
10

100
0

82

69

34
4
39

BE
# surgeons = 17

92

IT
13

1
19

14

123

132

3
21
81

JPN UK
57

10

11
31

19
19
16
23
50

111

171
115

14
20
16

FR

SP

GE ANZ CA SW

US

10

13

14

36

18

16

18

B
extreme,
very much

% of surgeons

moderate

some

none

# surgeons = 17

13

55

10

10

13

14

18

16

18

35

Figure 5. (A) Mean number of accesses placed by vascular access surgeons during training, by country. (B) Degree
of emphasis on vascular access creation during surgical training by country. Results are based on responses by
surgical operators responsible for creating new vascular accesses for hemodialysis patients in 222 facilities participating in Dialysis Outcomes and Practice Patterns Study II (DOPPS II; 2002-2004). Restricted to surgeons who created at
least 1 arteriovenous fistula (AVF) or graft (AVG) in the previous year. Abbreviations: ANZ, Australia and New Zealand;
BE, Belgium; CA, Canada; FR, France; GE, Germany; IT, Italy; JPN, Japan; SP, Spain; SW, Sweden; UK, United
Kingdom; US, United States; VA, vascular access. Adapted from Saran et al26 with permission of Wolters Kluwer
Health.

surgery trainees participated in access placements (AOR, 0.61; P 0.04).15 Reassuringly, 2

single-center studies in the United Kingdom
found that with proper supervision and allocation
of appropriate cases, fistula patency rates did not
differ after creation by trainees as opposed to
creation by senior consulting surgeons.29,30 Taken

in sum, these studies stress the importance of

adequate supervision during surgical training and
reinforce the critical impacts of experience and
judgment in creating successful fistulae.
Emphasis on type of access also is critical. In
the DOPPS, a remarkably high 54% of US access surgeons responded that the degree of em-

1038

phasis given to creating arteriovenous vascular

accesses was not at all emphasized or somewhat emphasized compared with other surgical
training (other possible responses were moderately emphasized, very much emphasized, or
extremely emphasized; Fig 5B).26 In contrast,
only 13% of the operators in Japan and 16% in
Italy, the nations with the highest prevalences of
fistulas, gave either of the 2 responses indicating
low training emphasis on access surgery.
The study of surgical background in the
DOPPS also showed considerable variability in
specialties of the operators.26 Vascular surgeons
predominated in Germany and Spain (92% of
facilities), whereas in the United States, 61% of
vascular access surgeries were conducted by
vascular surgeons, and 31% by general surgeons.
In Japan, there was an almost-equal split between nephrologists (29%), vascular surgeons
(29%), and urologists (28%), with general surgeons accounting for the remaining 14%. The
Italian experience is particularly interesting because nephrologists predominated in 85% of
facilities. In units in which a nephrologist created
the access, 100% of incident patients with a
permanent access had fistulas rather than AVGs.15
This observation suggests that sophisticated microsurgical dexterity may not be the key to
attaining optimal outcomes in HD hemoaccess,
but rather that devotion to creating the desired
arteriovenous access and meticulous emphasis
on the specific procedures associated with fistula
and AVG creation lead to success. One of the
authors of this report (F.L.) has decades of experience in training nephrologists in Italy and notes
that fellows at some institutions are mentored
carefully in the art and science of access creation
during their 5-year postgraduate specialty program. These trainees learn to complete a thorough preoperative clinical assessment, including
ultrasound mapping of the patients blood vessels, and a minimum of 2 months is devoted
exclusively to the teaching of interventional nephrology, with emphasis placed on creation of a
native radiocephalic anastomosis. The nephrologist-in-training typically masters this procedure
after 4-6 months of 2 surgeries per week and
masters mid- and proximal-forearm fistulas after
7-9 months. The most skilled and motivated
nephrology fellows also may learn to create
brachiocephalic fistulas or insert a synthetic AVG.

Goodkin et al
Box 1. Eleven FFBI Change Concepts for
Increasing Fistulas
1. Routine CQI review of vascular access
2. Timely referral to nephrologist
3. Early referral to surgeon for AVF only evaluation
and timely placement
4. Surgeon selection based on best outcomes, willingness, and ability to provide access services
5. Full range of surgical approaches to AV fistula
evaluation and placement
6. Secondary AV fistula placement in patients with AV
grafts
7. AV fistula placement in patients with catheters when
indicated
8. AV fistula cannulation training
9. Monitoring and maintenance to ensure adequate
access function
10. Education for caregivers and patients
11. Outcomes feedback to guide practice
Abbreviations: AV, arteriovenous; AVF, arteriovenous
fistula; CQI, continuous quality improvement; FFBI, Fistula
First Breakthrough Initiative.
Source: Fistula First website.2

Angioplasty may be performed in a procedure

room within the dialysis facility under local
anesthetic. Fellows also are trained in ultrasound
follow-up of vascular access and angioplastic
intervention and surgical correction procedures
for failing accesses. Unfortunately, during recent
years, fewer nephrologists are opting to acquire
the surgical skills needed for hemoaccess placement, in part because of increasing concerns with
malpractice litigation. Concurrently, the use of
catheters has increased and the use of fistulas has
begun to decrease in Italy (Fig 3). Whether this is
causal or coincidental is uncertain, although it
should be noted that there has been a concurrent
increase in age and burden of comorbid diseases
in the Italian dialysis population.
Follow the FFBI Change Concepts
The FFBI has called for 11 Change Concepts
to improve HD access outcomes, as listed in Box
1.2 Empirical evidence now supports several of
these recommendations. The first recommendation is for routine vascular access continuous
quality improvement. In an HD program, fistula
use increased from 15% to 43% and catheter use
was substantially limited after establishment of a
weekly multidisciplinary review meeting that
included nephrologists, vascular surgeons, interventional radiologists, and nurses.31

Vascular Access: Call to Arms

Late referrals to a nephrologist and vascular

surgeon during the course of chronic kidney
disease diminish the likelihood of attaining a
fistula (Change Concepts 2 and 3). In the DOPPS,
patients who received nephrologist care beginning more than 30 days before ESRD had almost
twice the likelihood of fistula use compared with
those who did not (AOR, 1.95; P 0.01).15 In
Europe, where fistula rates generally are higher,
84% of new HD patients had seen a nephrologist
more than 30 days before ESRD versus 74% in
the United States (P 0.0001). Similarly, a
study in Italy found that late referral was associated with a significant decrease in primary fistula
survival,32 whereas a study in Miami reported
that inadequate predialysis care explained almost
half the cases of patients starting HD therapy
using a catheter for access.33 DOPPS data also
have shown that the likelihood of having a permanent arteriovenous conduit for vascular access
(fistula or AVG) versus a percutaneous catheter is
markedly impeded by delays in attaining access
procedures: the AOR decreased 11% per 5-day
delay in median time from referral to evaluation
(P 0.0001) and decreased 6% per 5-day delay in
median time from evaluation until surgery (P
0.0001).21 The 2 nations with the highest proportions of fistula use had the shortest median times
from vascular access referral to surgery (Japan, 1
day from referral to evaluation and 5 days from
evaluation to surgery; and Italy, 2 days from
referral to evaluation and 3 days from evaluation
to surgery). There is significant variability in
health care systemrelated delays across countries. In a study evaluating practices in Italy,
Portugal, Turkey, and the United Kingdom, patients in the nations with longer delays until
surgical referral (particularly the United Kingdom) spend more time dialyzing through indwelling catheters and without fistulas.34 This likely is
modifiable; for example, in Australia, implementation of a multifaceted intervention that included employment of a vascular access nurse
coordinator and an algorithm to prioritize surgery improved the proportion of patients starting
HD therapy with a fistula from 56% to 75% (P
0.007) and halved the number of catheter-days.35
Concepts 3 through 5 in Box 1 stress the
criticality of referring HD patients to capable
surgeons who are devoted to the creation of

1039

functional fistulas whenever feasible, as discussed.

Fistula cannulation practices also may be important. An opinion-based guideline (Guideline 8
in the National Kidney Foundations KDOQI
[Kidney Disease Outcomes Quality Initiative]
dialysis access guidelines of 2000) counsels that
a new fistula should be allowed to mature for at
least 1 month, and ideally for 3 to 4 months, prior
to cannulation.36 However, median time to first
cannulation of a fistula varies greatly among
countries: the US median was the longest at 98
days, consistent with the KDOQI ideal recommendation, whereas median values were 25 days
in Japan and 27 days in Italy, 2 nations with far
higher rates of attaining functioning fistulas.37 In
aggregate, there was no significant difference for
the rate of fistula failure between those initially
cannulated in 15-28 days compared with 43-84
days, although first cannulation in 14 days or
fewer was associated with a 2.1-fold increased
risk of failure compared with fistulas that were
not used until after 14 days (P 0.006). It
appears that it is not essential to allow 3-4
months for fistula maturation at centers with
successful fistula practices.
Medications and Strategies to Sustain Access
Patency Are Needed
Few interventional studies have shown that
current drug therapy can sustain arteriovenous
access patency. One trial evaluating 84 HD patients indicated that dipyridamole, but not aspirin, decreased primary thrombosis of new
AVGs.38 Recently, the Dialysis Access Consortium randomly assigned 649 patients to dipyridamole plus aspirin therapy versus placebo and
noted that treatment was associated with significantly increased primary patency of new AVGs
at 1 year,39 although the clinical value of the
overall median improvement in graft survival of
just 6 weeks has been questioned.40 A casecontrol study of 60 incident patients found that
both folic acid and statin therapy were associated
with significantly higher fistula patency rates at 1
and 2 years.41 The largest randomized interventional trial to date, also from the Dialysis Access
Consortium, studied 877 patients and found that
clopidogrel failed to increase the percentage of
fistulas that became suitable for dialysis.19 Other
studies have noted higher risk of access failure

1040

with subcutaneous versus intravenous epoetin42

and with targeting a hematocrit of 42%.43 The
DOPPS, although limited by its observational
design, did not note an association between erythropoiesis-stimulating agent dose and increased
risk of vascular access thrombosis. It is hoped
that novel agents targeted to impede neointimal
hyperplasia and/or fibrosis may preserve the patency of dialysis AVGs, with encouraging results
from an animal study of paclitaxel to inhibit
neointimal proliferation.44

CONCLUSIONS
Worldwide clinical data establish a compelling case for maximizing the use of fistulas and
minimizing the use of catheters. Multiple statistically rigorous cohort studies have shown a robust
association of fistula use with longer patency and
fewer access complications versus AVGs, as well
as with the lowest risks of infection, hospitalization, and death versus AVGs and catheters. It is
particularly remarkable to note that the difference
in mortality between US and European HD patients, which is statistically significant after adjusting for a battery of demographic and comorbid
factors, is no longer significant when differences
in vascular access prevalence rates are taken into
account. Thus, the greater use of catheters/AVGs
and markedly lower use of fistulas in the United
States may be killing patients.
It is reasonable to acknowledge that not every
HD patient is suitable for creation of a fistula and
a small number of patients with permanent arteriovenous vascular access develop high-output
congestive heart failure.45 However, fistula use
has been associated with lower cardiovascular
mortality overall compared with catheter use.46
Furthermore, the greatly superior overall clinical
outcomes associated with fistula use, when examining differences within or between regions, justify making every effort to maintain fistulas in
most HD patients. It is highly unlikely that a
randomized controlled trial of fistula versus AVG
or catheter use will ever be conducted based on
the strength and uniformity of observational studies in the field. The consistently high rates of
fistula use in nations such as Japan, Italy, and
Germany show that the goal of widespread fistula use is attainable, and the progressive improvement in fistula prevalence in the United
States during the past 14 years shows that efforts

Goodkin et al

such as dedicated implementation of the FFBI

concepts can yield benefit.
Clinical data have substantiated several key
observations regarding the prospects of patients
dialyzing through a fistula: (1) the stated preference of the medical director and nurse manager
in a dialysis facility was associated very strongly
with the likelihood of fistula use; if the staff
wanted fistulas, more fistulas were attained; (2)
risk of primary fistula failure was 34% lower
when fistulas were placed by surgeons who had
created at least 25 fistulas during training; (3)
emphasis placed on fistula primacy during surgical training and number of fistulae created during
training were linked to the odds of fistula placement in practice; and (4) fistula use in a facility is
lower in the United States if surgical trainees
participate in access procedures. The Italian experience shows that dedication to selection and
care of fistulas coupled with extensive mentoring
during training has yielded great success even
when nephrologists, rather than vascular surgeons, perform fistula operations.
Hence, it is imperative that dialysis clinicians
insist on fistula prioritization and intervene to
avoid routine placement of catheters and AVGs.
Surgical training programs must teach trainees
the critical importance of meticulous fistula creation and the need to salvage fistulas, when
possible. Surgical residencies and fellowships
must supervise adequate numbers of proper vascular access procedures for every trainee. Senior
surgeons must actively oversee the surgeons in
training to rectify the current circumstance of
diminished odds of attaining a fistula at centers
in which trainees operate. Academic nephrologists must ensure that their surgical teaching
colleagues are aware of these needs and the
evidence that supports them. Nephrologists and
surgeons must insist that only individuals with
sufficient training, experience, and motivation
perform HD access procedures. This may be
facilitated by increased nephrologist involvement in the creation of fistulas,47 with newer
organizations, such as the American Society of
Diagnostic and Interventional Nephrology, publishing guidelines for training and certifying nephrologists in a number of vascular access interventions.48
More than 10 years ago, Hakim and Himmelfarb eloquently issued an HD access call to

Vascular Access: Call to Arms

action.49 They noted that most nephrologists

believed they were at the mercy of the vascular
surgeon(s) in their institution with regard to the
timeliness, interest, quality control and follow-up data of hemodialysis access procedures.
They urged nephrologists to become more proactive in the planning, management, and salvage of
fistulas. Efforts such as the FFBI seem to have
aided the cause. In light of the breadth and depth
of subsequent clinical findings, dialysis clinicians and access surgeons should unite in the call
to arms to establish the highest standards in
fistula attainment and HD access management.

ACKNOWLEDGEMENTS
The authors thank Dr Sabina Libardi for sharing her
knowledge of training practices for Italian nephrologists in
the field of vascular access.
Support: The DOPPS is administered by Arbor Research
Collaborative for Health and is supported by scientific
research grants from Amgen (since 1996), Kyowa Hakko
Kirin (since 1999, in Japan), Genzyme (since 2009), and
Abbott (since 2009), without restrictions on publications.
Financial Disclosure: Dr Goodkin has consulted for Affymax, AMAG Pharmaceuticals, Amgen, FibroGen, Keryx
Biopharmaceuticals, Registrat-MAPI, and Xenon Pharmaceuticals. Dr Pisoni has received speaker fees from Amgen,
Kyowa Hakko Kirin, and Vifor; has served as a consultant
for Pursuit Vascular; and has served on an advisory panel for
Merck. Dr Locatelli has served on advisory boards for
Affymax, Amgen-Dompe, Janssen-Cilag, Merck, and Roche
and serves on a safety committee for Sandoz. Drs Port and
Saran have served on a nephrology advisory board for
Amgen.

REFERENCES
1. National Kidney Foundation. KDOQI Clinical Practice Guidelines and Clinical Practice Recommendations:
2006 updates: vascular access. http://www.kidney.org/
Professionals/kdoqi/guideline_upHD_PD_VA/va_guide1.
htm. Accessed September 10, 2009.
2. Fistula First. AVFthe first choice for hemodialysis.
http://www.fistulafirst.org/. Accessed September 10, 2009.
3. FFBI. Prevalent US data. http://www.fistulafirst.org/
AboutFistulaFirst/FFBIData.aspx. Accessed July 14, 2010.
4. Young EW, Goodkin DA, Mapes DL, et al. The Dialysis Outcomes and Practice Patterns Study: an international
hemodialysis study. Kidney Int. 2000;57(suppl 74):S74-S81.
5. Pisoni RL, Gillespie BW, Dickinson DM, Chen K,
Kutner MH, Wolfe RA. The Dialysis Outcomes and Practice
Patterns Study (DOPPS): design, data elements, and methodology. Am J Kidney Dis. 2004;44(suppl 2):S7-S15.
6. Pisoni RL, Arrington CJ, Albert J, et al. Facility
hemodialysis vascular access use and mortality in countries
participating in DOPPS: an instrumental variable analysis.
Am J Kidney Dis. 2009;53:475-491.

1041
7. Dhingra RK, Young EW, Hulbert-Shearon TE, Leavey
SF, Port FK. Type of vascular access and mortality in U.S.
hemodialysis patients. Kidney Int. 2001;60:1443-1451.
8. Polkinghorne KR, McDonald SP, Atkins RC, Kerr PG.
Vascular access and all-cause mortality: a propensity score
analysis. J Am Soc Nephrol. 2004;15:477-486.
9. Pastan S, Soucie JM, McClellan WM. Vascular access
and increased risk of death among hemodialysis patients.
Kidney Int. 2002;62:620-626.
10. Held PJ, Brunner F, Odaka M, Garcia JR, Port FK,
Gaylin DS. Five-year survival for end-stage renal disease
patients in the United States, Europe, and Japan, 1982-1987.
Am J Kidney Dis. 1990;15:451-457.
11. Goodkin DA, Bragg-Gresham JL, Koenig KG, et al.
Association of comorbid conditions and mortality in hemodialysis patients in Europe, Japan, and the United States: the
Dialysis Outcomes and Practice Patterns Study (DOPPS).
J Am Soc Nephrol. 2003;14:3270-3277.
12. Lacson E Jr, Wang W, Lazarus JM, Hakim RM.
Change in vascular access and mortality in maintenance
hemodialysis patients. Am J Kidney Dis. 2009;54:912-921.
13. Bradbury BD, Chen F, Furniss A, et al. Conversion of
vascular access type among incident hemodialysis patients:
description and association with mortality. Am J Kidney Dis.
2009;53:804-814.
14. Combe C, Pisoni RL, Port FK, et al. Dialysis Outcomes and Practice Patterns Study (DOPPS): data on the use
of central venous catheters in chronic hemodialysis. Nephrologie. 2001;22:379-384.
15. Pisoni RL, Young EW, Dykstra DM, et al. Vascular
access use in Europe and the United States: results from the
DOPPS. Kidney Int. 2002;61:305-316.
16. Woods JD, Turenne MN, Strawderman RL, et al.
Vascular access survival among incident hemodialysis patients in the United States. Am J Kidney Dis. 1997;30:50-57.
17. Churchill DN, Taylor DW, Cook RJ, et al. Canadian
hemodialysis morbidity study. Am J Kidney Dis. 1992;19:
214-234.
18. US Renal Data System. USRDS 2009 Annual Data
Report: Atlas of Chronic Kidney Disease and End-Stage
Renal Disease in the United States. Bethesda, MD: National
Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases; 2009:266, 267, 341.
19. Dember LM, Beck GJ, Allon M, et al. Effect of
clopidogrel on early failure of arteriovenous fistulas for
hemodialysis: a randomized controlled trial. JAMA. 2008;
299:2164-2171.
20. Ohira S, Kon T, Imura T. Evaluation of primary
failure in AV-fistulae (early fistula failure). Hemodial Int.
2006;10:173-179.
21. Ethier J, Mendelssohn DC, Elder SJ, et al. Vascular
access use and outcomes: an international perspective from
the Dialysis Outcomes and Practice Patterns Study. Nephrol
Dial Transplant. 2008;23:3219-3226.
22. Goodkin DA, Chen K, Held PJ, et al. Factors influencing type of permanent vascular access at start of hemodialysis (DOPPS). Proceedings National Kidney Foundation Clinical Nephrology Meeting. 1999;187-188.
23. Young EW, Dykstra DM, Goodkin DA, Mapes DL,
Wolfe RA, Held PJ. Hemodialysis vascular access prefer-

1042
ences and outcomes in the Dialysis Outcomes and Practice
Patterns Study (DOPPS). Kidney Int. 2002;61:2266-2271.
24. Raynor HC, Besarab A, Brown WW, Disney A, Saito
A, Pisoni RL. Vascular access results from the Dialysis
Outcomes and Practice Patterns Study (DOPPS): performance against Kidney Disease Outcomes Quality Initiative
(K/DOQI) clinical practice guidelines. Am J Kidney Dis.
2004;44(suppl 2):S22-26.
25. Kulawik D, Sands JJ, Mayo K, et al. Focused vascular access education to reduce the use of chronic tunneled
hemodialysis catheters: results of a network quality improvement initiative. Semin Dial. 2009;22:692-697.
26. Saran R, Elder SJ, Goodkin DA, et al. Enhanced
training in vascular access creation predicts arteriovenous
fistula placement and patency in hemodialysis patients:
results from the Dialysis Outcomes and Practice Patterns
Study. Ann Surg. 2008;247:885-891.
27. Prischl FC, Kirchgatterer A, Brandsttter E, et al.
Parameters of prognostic relevance to the patency of vascular access in hemodialysis patients. J Am Soc Nephrol.
1995;6:1613-1618.
28. Choi KL, Salman L, Krishnamurthy G, et al. Impact
of surgeon selection on access placement and survival following preoperative mapping in the Fistula First era. Semin
Dial. 2008;21:341-345.
29. Weale AR, Barwell J, Chant H, Lear PA, Mitchell
DC. The impact of training on outcomes in primary vascular
access surgery. Ann R Coll Surg Engl. 2004;86:275-280.
30. Gundevia Z, Whalley H, Ferring M, Claridge M,
Smith S, Wilmink T. Effect of operating surgeon on outcome
of arteriovenous fistula formation. Eur J Vasc Endovasc
Surg. 2008;35:614-618.
31. Vassalotti JA, Falk A, Teodorescu V, Uribarri J. The
multidisciplinary approach to hemodialysis vascular access
at the Mount Sinai Hospital. Mt Sinai J Med. 2004;71:94102.
32. Ravani P, Brunori G, Mandolfo S, et al. Cardiovascular comorbidity and late referral impact arteriovenous fistula
survival: a prospective multicenter study. J Am Soc Nephrol.
2004;15:204-209.
33. Lenz O, Sadhu S, Fornoni A, Asif A. Overutilization
of central venous catheters in incident hemodialysis patients: reasons and potential resolution strategies. Semin
Dial. 2006;19:543-550.
34. DiBenedetto A, Basci A, Cesare S, Marcelli D, Ponce
P, Richards N. Increased use of catheters as vascular access:
is it justified by patients clinical conditions? J Vasc Access.
2007;8:21-27.
35. Polkinghorne KR, Seneviratne M, Kerr PG. Effect of
a vascular access nurse coordinator to reduce central venous

Goodkin et al
catheter use in incident hemodialysis patients: a quality
improvement report. Am J Kidney Dis. 2009;53:99-106.
36. National Kidney Foundation. KDOQI 2000 Guidelines
for Vascular Access. http://www.kidney.org/professionals/kdoqi/
guidelines_updates/doqiupva_i.html#doqiupva8. Accessed October 6, 2009.
37. 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
Int. 2003;63:323-330.
38. Sreedhara R, Himmelfarb J, Lazarus JM, Hakim RM.
Anti-platelet therapy in graft thrombosis: results of a prospective, randomized, double-blind study. Kidney Int. 1994;45:
1477-1483.
39. Dixon BS, Beck GJ, Vazquez MA, et al; for the DAC
Study Group. Effect of dipyridamole plus aspirin on hemodialysis graft patency. N Engl J Med. 2009;360:2191-2201.
40. Lok CE. Antiplatelet therapy and vascular-access
patency. N Engl J Med. 2009;360:2240-2242.
41. Righetti M, Ferrario G, Serbelloni P, Milani S, Tommasi A. Some old drugs improve late primary patency rate of
native arteriovenous fistulas in hemodialysis patients. Ann
Vasc Surg. 2009;23:491-497.
42. Lee Y-K, Koo J-R, Kim J-K, et al. Effect of route of
EPO administration on hemodialysis arteriovenous vascular
access failure: a randomized controlled trial. Am J Kidney
Dis. 2009;53:815-822.
43. Besarab A, Bolton WK, Browne JK, et al. The effects
of normal as compared with low hematocrit values in patients with cardiac disease who are receiving hemodialysis
and epoetin. N Engl J Med. 1998;339:584-590.
44. Kelly B, Melhem M, Zhang J, et al. Perivascular
paclitaxel wraps block arteriovenous graft stenosis in a pig
model. Nephrol Dial Transplant. 2006;21:2425-2431.
45. Amerling R. Fistula first: is it time to curb our
enthusiasm? Nephrol News Issues. 2006;20(9):38-39.
46. Wasse H, Speckman RA, McClellan WM. Arteriovenous fistula use is associated with lower cardiovascular
mortality compared with catheter use among ESRD patients.
Semin Dial. 2008;21:483-489.
47. Mishler R. Autologous arteriovenous fistula creation
by nephrologists. Adv Chronic Kidney Dis. 2009;16:321328.
48. Asif A, Besarab A, Roy-Chaudhury P, Spergel LM,
Ravani P. Interventional nephrology: from episodic to
coordinated vascular access care. J Nephrol. 2007;20:399405.
49. Hakim R, Himmelfarb J. Hemodialysis access failure: a call to action. Kidney Int. 1998;54:1029-1040.