Surg Clin N Am 87 (2007) 1213–1228

Successful Angioaccess
Niten Singh, MDa,*, Benjamin W. Starnes, MD, FACSb,
Charles Andersen, MDa
a
Vascular and Endovascular Surgery Madigan Army Medical Center,
9040-A Fitzsimmons Avenue, Tacoma, WA 98431, USA
b
Vascular and Endovascular Surgery, Harborview Medical Center,
University of Washington, Seattle, WA, USA

Surgery for hemodialysis (HD) access is the most commonly performed

vascular surgical operation in the United States, predominantly because
of a steady increase in the prevalence of end-stage renal disease (ESRD).
Despite a concomitant increase in the mean age of these patients and more
coexisting morbidities, advances in the management of renal failure and
dialysis have allowed for a longer survival among patients on HD. The
Achilles heel for these patients remains vascular access, however, with
poor long-term patency rates resulting in multiple interventions for throm-
bosis and maintenance, and, in many patients, the eventual need for lifelong
catheter placement. Access failure is the second leading cause of hospitaliza-
tion among patients who have ESRD, and the annual cost of access main-
tenance is estimated to be $1 billion in the United States [1].
Multiple studies have confirmed the improved patency rate and lower
infection rates for native arteriovenous fistulae (AVF) compared with pros-
thetic arteriovenous grafts (AVG). Although approximately 60% to 90%
of nonautogenous grafts are functional at 1 year, the patency falls to 40%
to 60% at 3 years [2]. Despite inferior patency rates, prosthetic grafts con-
tinue to be more common than native fistulae in the United States, account-
ing for 65% of all access procedures. In contrast, data from Canada and
Europe demonstrate greater use of autogenous fistulae in those countries,
with prosthetic grafts accounting for less than 35% of all access procedures
in Canada and 10% in Europe [3].
In an effort to promote more standardized practice patterns and greater
success with autogenous access placement, the National Kidney Foundation

* Corresponding author.
E-mail address: nhsingh@aol.com (N. Singh).

0039-6109/07/$ - see front matter. Published by Elsevier Inc.

doi:10.1016/j.suc.2007.08.004 surgical.theclinics.com
1214 SINGH et al

introduced the Dialysis Outcome Quality Initiative (DOQI) [4]. In the past
these guidelines set the goal of greater than 50% of all dialysis access being
autogenous. The most recent recommendation is a prevalence of AVF of
greater 65% by 2009 [5]. This performance metric is a goal that the Center
for Medicare Services expects for reimbursement. Although it is a useful
guideline, it does not take into account variations in patient populations.
Many patients are not referred to a surgeon per DOQI guidelines when their
creatinine clearance is less than 25 mL/min for access consideration before
initiation of HD. Many patients present with a tunneled catheter in place
that is malfunctioning and may not have time to wait for an autogenous
access to mature. Usually it is between these two extremes that we, as
surgeons, are asked to see these patients. Using a ‘‘fistula first’’ practice re-
sults in better long-term patency but must be tempered with the needs of the
specific patient in question.
In formulating a strategy for successful dialysis access a comprehensive
approach should be undertaken. This approach involves understanding
this patient population and working with the nephrologist to create an
environment that is beneficial to the patient. Surgeons who perform these
procedures understand that the challenge is not the simple technical process of
creating an AVF, but understanding the potential pitfalls that can be associated
with access. The preoperative planning, as with any surgical procedure, is the
most important aspect, followed by the postoperative maintenance of the
access.

Preoperative strategy
The planning for HD access should ideally allow for adequate matura-
tion time for an AVF, which is between 6 and 8 weeks (up to 12 weeks).
This timing would optimize the use of autogenous access; however, this is
often not possible and the patient requires access sooner. Every surgeon
should adopt a standard algorithm for HD access. Our algorithm of access
is as follows: nondominant upper extremity, the most distal site first, and
autogenous access before prosthetic. Within every algorithm there exist
variations. For example, women are more likely to undergo insertion of
prosthetic grafts than men and have higher failure rates of AVF, which is
likely a reflection of the larger caliber of superficial arm veins in men [6].
Maintaining a standard practice has been shown to increase the percentage
of autogenous access, however [7].
History should include any prior access procedures or recent intravenous
lines, particularly central lines, because subclavian lines are associated with
an extremely high rate of stenosis [8]. Physical examination should focus on
bilateral upper extremity blood pressure to detect a potential subclavian ar-
tery stenosis and palpation of the brachial and radial pulses. An Allen test
should be performed to ensure there is adequate radial and ulnar flow to the
hand (Fig. 1).
 SUCCESSFUL ANGIOACCESS 1215

Fig. 1. Allen test.

Ultrasound may be incorporated into preoperative planning, particularly

for autogenous access placement. Duplex ultrasound (DU) has been used
for preoperative planning in dialysis access, and studies have shown a proven
benefit of ultrasound in predicting success in patients who have undergone
a preoperative ultrasound [9]. Robbin and colleagues [10] noted an increase
in autogenous arteriovenous access creation from 32% to 58% after a preop-
erative DU program was started. With preoperative ultrasonography, Silva
and colleagues [11] were also able to demonstrate an increase in autogenous
access placement from 14% to 63% and a decrease in failure of autogenous
access from 38% to 8.3%. In that study, veins greater than 2.5 mm were
required for autogenous access placement and greater than 4 mm for non-
autogenous placement. Using the criteria of a 2-mm vein at the wrist and
greater than 3 mm in the upper arm, Ascher and colleagues [12] reported
a similar increase in autogenous access placement with preoperative DU.
Most preoperative ultrasound procedures can be performed in an office
or vascular laboratory setting. The forearm venous network is superficial
and easily imaged (Fig. 2). Superficial veins can be visualized longitudinally
and in cross section. If these veins are compressible, greater than 2.5 mm, and
patent throughout their course, then an autogenous access can likely be per-
formed. Mendes and colleagues [13] studied the cephalic vein preopera-
tively with DU to determine whether a minimal cephalic vein size in the
forearm could predict successful wrist autogenous access. They noted
that patients who had a cephalic vein size of 2.0 mm or smaller were
less likely to have a successful wrist autogenous access than if the cephalic
vein was greater than 2.0 mm [13]. If the patient’s history suggests previ-
ous central venous lines or arm swelling, the deep veins, such as the
1216 SINGH et al

Fig. 2. Cephalic vein ultrasound.

axillary and distal subclavian, can also be interrogated (Fig. 3). The bony
clavicle limits accurate assessment at the mid to proximal subclavian level.
Color flow DU may also be used to assess the arterial inflow. Starting at
the wrist the radial artery can be identified and followed proximally to the
brachial artery. Using color flow in the longitudinal view, occlusive arterial
disease can be identified. Unobstructed arterial inflow of 2.0 mm has been
used as a predictor of success [14,15].
Arteriography and venography may be incorporated into the preopera-
tive planning of patients. Iodinated contrast agents are nephrotoxic and car-
bon dioxide or gadolinium should be used instead in patients who have
residual renal function. Arteriography is usually recommended in patients
who have evidence of significant inflow disease as directed by physical exam-
ination or ultrasonography. If a short-segment subclavian stenosis is identi-
fied, it may be treated with angioplasty and stenting. Contrast venography
should be incorporated in the preoperative planning in patients who have
poor veins noted on duplex or if there is a history of prior central venous
access, particularly in the subclavian vein [16,17].
After adequate preoperative planning the choice of access can be deter-
mined and can either be autogenous or prosthetic. Generally in our practice,
when patients are referred relatively early we aggressively attempt
 SUCCESSFUL ANGIOACCESS 1217

Fig. 3. Ultrasound of proximal subclavian vein.

autogenous access and attempt radiocephalic access with a smaller cephalic

vein (ie, 1.6 mm). The reason for this is that it is a relatively benign proce-
dure and potentially has the benefit of increasing the caliber of the more
proximal veins if it does not mature itself.

Autogenous access
The first choice of autogenous access is the radial artery to cephalic vein
fistula. First described in 1966, it is the most well-known autogenous access
performed and is also referred to as the Brescia-Cimino fistula or the wrist
fistula [18]. It is performed in various ways but the most common involves
an end of the cephalic vein to the side of the radial artery. This procedure
can be accomplished with one or two incisions. In general the vein is mobi-
lized near the wrist, an adequate segment of the radial artery is identified,
and the anastomosis is performed with 6-0 polypropylene suture in a running
style (Fig. 4). This fistula is then given 6 to 8 weeks to mature before access-
ing it for HD. Many times this AVF is slow to mature because large
tributaries in the forearm may shunt flow away from the cephalic vein.

Fig. 4. Radiocephalic fistula.

1218 SINGH et al

Identification of these tributaries by physical examination, ultrasonography,

or fistulogram and ligation usually allows the cephalic vein to subsequently
mature.
Another variation of the Brescia-Cimino fistula is the ‘‘snuffbox’’ fistula
[19]. It is an anastomosis between the distal cephalic vein and the thenar
branch of the radial artery that can be found in the anatomic snuffbox of
the hand (Fig. 5). It can generally be performed through one incision over
the snuffbox and is the most distal autogenous fistula. It is an end of the ce-
phalic vein to the side of the thenar branch of the radial artery anastomosis.
It may require finer suture (7-0) because the vessels are smaller. This fistula
affords the luxury of an extremely small incision and the potential of
increasing the size of the more distal veins.
Another forearm autogenous access is the radial artery to forearm basilic
vein transposition [20]. It involves an incision along the forearm portion of

Fig. 5. Snuffbox fistula. (From Hallett JW Jr, Mills JL, Earnshaw JJ, et al, editors. Comprehen-
sive vascular and endovascular surgery. St. Louis: Mosby; 2004; with permission.)
 SUCCESSFUL ANGIOACCESS 1219

the basilic vein and mobilization toward the antecubital fossa. The radial
artery is than exposed and a subcutaneous tunnel is created to allow the
basilic vein to be transposed toward the proximal radial artery (Fig. 6). Us-
ing this fistula still preserves the upper arm access if it does not mature.
The brachial artery to cephalic vein fistula is the next higher-level autoge-
nous fistula. It involves an anastomosis between the brachial artery and
cephalic vein in the antecubital fossa. It is generally performed in an end of
cephalic vein to side of brachial artery configuration. It has excellent flow

Fig. 6. Forearm basilic vein transposition. (From Hallett JW Jr, Mills JL, Earnshaw JJ, et al, ed-
itors. Comprehensive vascular and endovascular surgery. St. Louis: Mosby; 2004; with permission.)
1220 SINGH et al

and maturation rates but has been associated with higher rates of ‘‘steal’’ phe-
nomenon [21]. It also eliminates the forearm for consideration of future access.
The brachial artery to basilic vein fistula is the last autogenous conduit in
the upper arm. The operation is done under general anesthesia; it requires
more extensive dissection because the basilic vein is deeper in the upper
arm. The preoperatively marked basilic vein requires transposition to
a more superficial location so it can be accessed. Results have varied with
some reporting excellent patency and others reporting less favorable results
[22,23].

Prosthetic access
AVGs are associated with poorer patency than an AVF; however, they do
have the advantage of easier cannulation and a much shorter time to matura-
tion requiring only 14 days before access can be achieved. Generally these
grafts are made from expanded polytetrafluoroethylene (PTFE) and come
in a variety of sizes. Most grafts are 6 mm but there are also tapered grafts
with a 4-mm arterial end and a 7-mm venous end. The principle cause of fail-
ure in an AVG (80%) is neointimal hyperplasia at the venous anastomosis.
Compared with AVF, however, the secondary patency rates may be higher
because salvage procedures are more readily performed in prosthetic grafts.
The most distal forearm graft is the straight radial artery to cephalic or an-
tecubital vein graft. Another forearm graft is the brachial to cephalic or basilic
vein loop graft. These grafts are performed with end of graft to side of artery
and vein anastomosis with 6-0 polypropylene sutures or CV6 Gore-Tex PTFE
suture. Again these AVGs provide relatively simple access and may be used in
the face of disadvantaged forearm veins (Figs. 7 and 8).
The next level of prosthetic is the brachial artery to axillary vein graft
that provides the last level of access in the upper arm. It can be performed
after a failed brachial-basilic AVF as long as the axillary vein is patent and it
is an end of graft to side of artery and vein anastomosis. Other potential and
less often used sites are the cross-chest axillary artery to axillary vein graft
and lower extremity thigh loop grafts, which are not discussed here.

Complications
Failure of maturation of arteriovenous fistulae
AVFs may take up to 12 weeks to mature but in an access such as the
radial-cephalic fistula the cause may be multiple tributaries of the cephalic
vein that should be ligated. This ligation can be accomplished by identifying
prominent vessels on physical examination, using duplex ultrasonography
to mark these, or using fistulograms to identify and ligate these branches.
Even if the radiocephalic AVF does not mature properly it may promote
 SUCCESSFUL ANGIOACCESS 1221

Fig. 7. Prosthetic straight arm grafts. (From Hallett JW Jr, Mills JL, Earnshaw JJ, et al, editors.
Comprehensive vascular and endovascular surgery. St. Louis: Mosby; 2004; with permission.)

the enlargement of the more prominent veins to allow for a more durable
proximal upper extremity fistula.

The failing or thrombosed access

Often a patient is referred back to the surgeon with a question of a poorly
functioning AVF and AVG. Unlike AVGs, an AVF can remain patent with
1222 SINGH et al

Fig. 8. Prosthetic loop grafts. (From Hallett JW Jr, Mills JL, Earnshaw JJ, et al, editors. Compre-
hensive vascular and endovascular surgery. St. Louis: Mosby; 2004; with permission.)

minimal flow, therefore Doppler or duplex interrogation of a suspected

thrombosed AVF should be performed to confirm or exclude the diagnosis.
High recirculation times, elevated venous pressures, and inability to achieve
adequate urea clearance are signs of a venous outflow obstruction. The
treatment of a thrombosed AVG has been described in several series and
can range from endovascular thrombolysis to surgical thrombectomy with
 SUCCESSFUL ANGIOACCESS 1223

patch angioplasty of the venous outflow. Rescue of a failing or thrombosed

AVF has been mixed with AVG studies but in general they are not as easily
salvaged. These results are similar to a thrombosed autogenous lower
extremity bypass [24,25].
Endovascular therapy involves assessing the venous outflow and dilating
any stenosis noted along with a central venogram to rule out a central vein
stenosis. One technique that is used for thrombosed AVGs is the ‘‘lyse and
wait’’ technique. This technique involves cannulating the AVG with an
18-guage Angiocath and instilling 3 mg of tissue plasminogen factor (tPA)
and 3000 units of heparin while compressing the arterial and venous ends
of the graft. After waiting anywhere from 30 minutes to 2 hours a venogram
is performed. Any outflow stenosis is angioplastied. Next a sheath, directed
toward the arterial end, is placed and a Fogarty embolectomy catheter is
used to dislodge the arterial plug. This technique is useful and can be
repeated without having to perform surgical correction or revision and
unidentified lesions can be found and treated from the venous outflow tract
to the central veins [26].
Surgical thrombectomy is fairly standard with the incision placed over
the venous outflow and the use of a Fogarty catheter to remove the arterial
thrombus. The venous end then has a PTFE or Dacron patch placed. Com-
pletion venography should be performed to visualize the venous outflow and
identify lesions that may require further treatment.

Venous hypertension
Mild arm swelling is a common finding in these patients; however, this
can be a more devastating complication with some patients having severe
swelling. This complication is usually the result of outflow vein or subcla-
vian or central vein stenosis or occlusion. This problem can be treated by
angioplasty of the central vein with good initial results; however, these are
prone to recurrence and the need for reintervention [27]. If swelling is not
controlled and the patient begins to have signs of venous ulceration, then
ligation of the AVF or AVG is warranted.

Infection
AVGs are more prone to infection, particularly with Staphylococcus
aureus. The very nature of a superficial graft, with repeated punctures in per-
haps not the most sterile of environments, and the impaired immunity of these
dialysis patients all play a role [28]. If a graft is suspected of infection without
any signs of exposed graft than a trial of antibiotics should be initiated. If
there is a tract or exposed graft then excision of this portion of the graft should
be performed with removal of all of the unincorporated graft. If it is a short
section, using another graft tunneled in a clean plane between the uninfected
incorporated portions of the graft can be performed (Fig. 9) [29].
1224 SINGH et al

Fig. 9. Segmental excision and bypass of infected graft. (From Hallett JW Jr, Mills JL,
Earnshaw JJ, et al, editors. Comprehensive vascular and endovascular surgery. St. Louis:
Mosby; 2004; with permission.)

Pseudoaneurysm
Pseudoaneurysm (PSA) formation within an AVF does not necessarily
imply infection, because repeated cannulation can cause this problem. A
PSA of an AVG may be associated with infection because the scar tissue
around the graft limits the formation of a PSA. If a PSA continues to en-
large and the overlying skin is threatened, it should be repaired. This repair
can be done with segmental excision and bypass or with emerging endovas-
cular techniques with a covered stent. There are case reports describing this
technique and the results have been variable [30].
Steal
Creation of either an AVF or AVG may predispose a patient to ischemic
steal symptoms. This phenomenon may manifest as cool digits or profound
ischemic changes, such as tissue loss. A patient develops steal because of the
reduced pressure just proximal to the fistula secondary to the large capaci-
tance of the venous outflow and blood preferentially flowing in this direc-
tion. Generally this can be managed with a trial of observation if the
symptoms are mild. To diagnose steal various techniques from physical
examination to arteriography are required. If compression of the AVF
results in restoration of a palpable radial pulse and resolution of symptoms,
a confirmatory arteriogram to rule out an inflow stenosis or more distal
native arterial stenosis is required. The brachial artery to cephalic vein
fistula is the fistula most commonly associated with this complication.
The treatment can involve revision of the anastomosis or operative banding,
which has not been too successful. Another procedure that involves preser-
vation of the access site is the distal revascularization with interval ligation
(DRIL) procedure. Originally described by Shanzer and colleagues [31] it in-
volves ligating the artery distal to the fistula and than constructing a bypass
3 to 5 cm above the AVF anastomosis (above the reduced pressure area) to
the distal native artery (Fig. 10). The DRIL procedure is extremely effective
 SUCCESSFUL ANGIOACCESS 1225

Fig. 10. DRIL procedure. (From Hallett JW Jr, Mills JL, Earnshaw JJ, et al, editors. Compre-
hensive vascular and endovascular surgery. St. Louis: Mosby; 2004; with permission.)

and should be used in patients who have a functioning native AVF and lim-
ited access availability. The decision to ligate an axial artery for preservation
of a prosthetic fistula needs to be weighed carefully with consideration given
to the lifespan of a prosthetic graft. Another technique is proximalization of
the arterial anastomosis, which has recently been described with good re-
sults by Zanow and colleagues [32]. This technique involves dissecting the
arterialized fistula vein near the AVF anastomosis and ligating and dividing
it. The inflow artery (more proximal brachial or axillary artery) is then dis-
sected and a 4-mm expanded PTFE graft is used between this artery and the
efferent vein to construct a bypass (Fig. 11). In their series 84% of access-re-
lated ischemic symptoms resolved. The increasing use of endovascular pro-
cedures has led to novel approaches to difficulties with HD access. Recently
1226 SINGH et al

Fig. 11. Proximalization of arterial inflow. Arterial anastomosis divided and a tapered graft
placed between the more proximal axillary artery and outflow (cephalic) vein. (From Zanow J,
Kruger U, Scholz H. Proximalization of the arterial inflow: a new technique to treat access-related
ischemia. J Vasc Surg 2006;43:1217; with permission.)

steal has been managed using minimally invasive techniques whereby the ve-
nous end of the AV anastomosis is accessed with a wire and a 4-mm angio-
plasty balloon inflated. A small incision is made over the vein and a single
Prolene suture tied around the vein with the balloon inflated. The vein is
thus banded with exacting fashion to relieve steal. With this technique,
the authors have noted a 94% technical success rate for treating patients
who have steal, thus avoiding an extensive DRIL procedure [33].

Summary
Creating effective dialysis access requires a comprehensive plan formu-
lated between the nephrologist, the surgeon, and the patient. In this setting,
HD access is the patient’s lifeline and maintaining it requires a plan to deal
with the aforementioned complications and sometimes low patency rates.
To optimize success, the most important aspect, as with any other surgical
procedure, is the preoperative planning. Understanding and dealing with the
potential complications allows the surgeon the ability to optimize the prac-
tice of HD access.

References
[1] United States Renal Data System. The economic cost of ESRD, vascular access procedures,
and Medicare spending for alternative modalities of treatment. Am J Kidney Dis 1997;30:
S160–77.
 SUCCESSFUL ANGIOACCESS 1227

[2] Schwab S, Harrington J, Singh A, et al. Vascular access for hemodialysis. Kidney Int 1999;
55:2078–90.
[3] Dixon BS, Novak L, Fangman J. Hemodialysis vascular access survival: upper-arm native
arteriovenous fistula. Am J Kidney Dis 2002;39:92–101.
[4] NKF-K/DOQI clinical practical guidelines for vascular access: update 2000. Am J Kidney
Dis 2001;29:223–9.
[5] NKF-K/DOQI clinical practical guidelines for vascular access: update 2006. Am J Kidney
Dis 2006;48(Suppl 1):S258–63.
[6] Ifudu O, Macey LJ, Homel P, et al. Determinants of type of initial hemodialysis vascular
access. Am J Nephrol 1997;17:425–7.
[7] Kalman PG, Pope M, Bhola C, et al. A pratical approach to vascular access for hemodialysis
and predictors of success. J Vasc Surg 1999;30:727–33.
[8] Spinowitz BS, Galler M, Golden RA, et al. Subclavian vein stenosis as a complication of
subclavian vein catheterization for hemodialysis. Arch Intern Med 1987;147:305–7.
[9] Koskoy C, Kuzu A, Erden I. Predictive value of color Doppler ultrasonography in detecting
failure of vascular access grafts. Br J Surg 1995;82:50–2.
[10] Robbin ML, Gallichio MH, Deierhoi MH, et al. US vascular mapping before hemodialysis
placement. Radiology 2000;217:83–8.
[11] Silva MB, Hobson RW, Pappas PJ, et al. A strategy for increasing the use of autogenous
hemodialysis access procedures: impact of preoperative noninvasive evaluation. J Vasc
Surg 1998;27:302–8.
[12] Ascher E, Gade P, Hingorani A, et al. Changes in the practice of angioaccess surgery: impact
of the dialysis outcome and quality initiative recommendations. J Vasc Surg 2000;31:84–92.
[13] Mendes RR, Farber MA, Marston WA, et al. Prediction of wrist arteriovenous fistula mat-
uration with preoperative vein mapping with ultrasonography. J Vasc Surg 2002;36:460–3.
[14] Parmley MC, Broughan TA, Jennings WC. Vascular ultrasonography prior to dialysis
access surgery. Am J Surg 2002;184:568–72.
[15] Malovrh M. The role of sonography in the planning of arteriovenous fistulas for hemodial-
ysis. Semin Dial 2003;16:299–303.
[16] Tordoir JH, de Bruin HG, Hoeneveld H, et al. Duplex ultrasound scanning in the assessment
of arteriovenous fistulas created for hemodialysis access: comparison with digital subtraction
angiography. J Vasc Surg 1989;10:122–8.
[17] Glanz S, Bashit B, Gordon DH, et al. Axillary and subclavian vein stenosis: percutaneous
angioplasty. Radiology 1988;168:371–3.
[18] Brescia M, Cimino J, Appel K, et al. Chronic hemodialysis using venipuncture and surgically
created arteriovenous fistula. N Engl J Med 1966;275:1089–92.
[19] Wolowczyk L, Williams AJ, Donovan KL, et al. The snuffbox arteriovenous fistula for
vascular access. Eur J Vasc Endovasc Surg 2000;19:70–6.
[20] Silva MB, Hobson RW, Pappas PJ, et al. Vein transposition in the forearm for autogenous
hemodialysis access. J Vasc Surg 1997;26:981–8.
[21] Sparks SR, Vanderlinden JL, Gnanadev DA, et al. Superior patency of perforating antecu-
bital vein arteriovenous fistulae for hemodialysis. Ann Vasc Surg 1997;11:165–7.
[22] Coburn MC, Carney WI. Comparison of basilic vein and polytetrafluoroethylene for
brachial arteriovenous fistula. J Vasc Surg 1994;20:896–902.
[23] Taghizadeh A, Dasgupta P, Khan MS, et al. Long-term outcomes of brachiobasilic transpo-
sition fistula for haemodialysis. Eur J Vasc Endovasc Surg 2003;26:670–2.
[24] Lay JP, Ashleigh RJ, Tranconi L, et al. Results of angioplasty of Brescia-Cimino haemodial-
ysis fistulae: medium term follow-up. Clin Radiol 1998;53:608–11.
[25] Safa AA, Vaji K, Roberts A, et al. Detection and treatment of dysfunctional hemodialysis
access grafts: effects of surveillance program on graft patency and the incidence of thrombo-
sis. Radiology 1996;199:653–7.
[26] Cynamon J, Lakritz P, Wahl S, et al. Hemodialysis graft declotting: description of the ‘‘lyse
and wait’’ technique. J Vasc Interv Radiol 1997;8:825–9.
1228 SINGH et al

[27] Lumsden AB, MacDonald MJ, Isiklar H, et al. Central venous stenosis in the hemodialysis
patient: incidence and efficacy of endovascular treatment. Cardiovasc Surg 1997;5:504–9.
[28] Tokars JI, Light P, Anderson J, et al. A prospective study of vascular access infections at
seven outpatient hemodialysis centers. Am J Kidney Dis 2001;37:1232–40.
[29] Schwab DP, Taylor SM, Cull DL, et al. Isolated arteriovenous dialysis access graft segment
infection: the results of segmental bypass and partial graft excision. Ann Vasc Surg 2000;14:
63–6.
[30] Ryan JM. Using a covered stent (Wallgraft) to treat pseudoaneurysms of dialysis grafts and
fistulas. Am J Radiol 2002;180:1067–71.
[31] Schanzer H, Skadany M, Haimov M. Treatment of angioaccess-induced ischemia by revas-
cularization. J Vasc Surg 1992;16:861–6.
[32] Zanow J, Kruger U, Scholz H. Proximalization of the arterial inflow: a new technique to treat
access-related ischemia. J Vasc Surg 2006;43:1216–21.
[33] Riley JT, Miller GA, Wu T, et al. Minimally invasive limited ligation endoluminal revision
(MILLER) of arteriovenous fistulas to correct steal syndrome: results and clinical outcomes
at one year. Presented at the 2007 Vascular Annual Meeting, Society for Vascular Surgery.
Baltimore, MD, June 10, 2007.