J Artif Organs
DOI 10.1007/s10047-016-0927-4
ORIGINAL ARTICLE
Axillo-iliac arteriovenous hemodialysis graft creation
with an early cannulation device
Carlos A. Hinojosa1 • Javier E. Anaya-Ayala1 • Alejandra Lopez-Mendez1
Zeniff Gomez-Arcive1 • Hugo Laparra-Escareno1 • Cesar Cuen-Ojeda1 •
Rene Lizola1 • Adriana Torres-Machorro1
Received: 13 July 2016 / Accepted: 27 September 2016
Ó The Japanese Society for Artificial Organs 2016
Abstract Exhaustion of superficial veins coupled with the
presence of intrathoracic central venous occlusions remains a
significant obstacle for hemodialysis access creation; com-
plex arteriovenous graft (AVG) configurations have been
described. The axillary-iliac AVG was first reported in 1987,
and few authors have explored this access. We evaluated our
experience with this AVG configuration utilizing the early
cannulation (EC) graft FlixeneTM (Atrium TM, Hudson, NH,
USA). Eight patients (75 % men; mean age 37 ± 10 years)
with End-Stage Renal Disease (ESRD) underwent axillo-il-
iac AVG creation with FlixeneTM grafts; all had exhausted
peripheral veins, occluded thoracic central veins, and inad-
equate femoral veins. Inflow from the axillary artery and
outflow in iliocaval system was assessed prior to access
creation. An axillary-to-common iliac AVG was constructed
using a 6 mm (mm) EC graft and tunneled in the chest and
abdominal wall. Eight grafts were implanted; all were patent
after placement. Seven (88 %) were successfully used for
hemodialysis within 72 h and one (12 %) within 96. During
the mean follow-up of 6 months, 5 (62 %) patients under-
went thrombectomy, 1 (12 %) of them had balloon angio-
plasty at the vein anastomosis, and 2 (25 %) grafts were
removed secondary to infection. The remaining grafts are
still functioning. Complications as high-output heart failure,
steal syndrome and venous hypertension were not observed.
Construction of axillo-iliac AVG with EC grafts in the setting
of exhausted veins, occluded intrathoracic central veins and
& Carlos A. Hinojosa
carlos.a.hinojosa@gmail.com
1
Department of Surgery, Section of Vascular Surgery and
Endovascular Therapy, Instituto Nacional de Ciencias
Médicas y Nutrición ‘‘Salvador Zubirán’’, Vasco De Quiroga
15, Tlalpan, Sección XVI, 14080 Mexico City, Mexico
Blood Vessel Prosthesis
•
hostile groins, is a viable arteriovenous access alternative
while avoiding central venous catheters.
Keywords Axillo-iliac arteriovenous grafts
Early
cannulation grafts
Technique
Background
Exhaustion of suitable superficial veins in upper extremities
for hemodialysis access creation coupled with the presence
of intrathoracic central venous occlusion is a significant
problem in end-stage renal disease (ESRD) patients; [1, 2]
complex and exotic arteriovenous graft (AVG) configura-
tions have been proposed and described to overcome the
anatomic and clinical challenges of this patient population
[1–3]. The construction and availability of such complex
access procedures are limited not only to the presence of
adequate arterial inflow and venous outflow, but also to the
creativeness and experience of vascular access surgeons.
The axillary artery to iliac vein vascular access tunneled
under the chest and abdominal wall was first reported in 1987
by Cimochowski et al. [4], and the authors implanted 9
polytetrafluoroethylene (PTFE) grafts in 8 patients with
technical success in all cases and acceptable functionality at
12 months. Since then, this technique has been described in a
limited fashion and few reports using this specific configu-
ration of AVGs are available in the literature [5–7].
Methods
We evaluated retrospectively our experience with the
axillo-iliac AVG using the early cannulation (EC) graft
FlixeneTM (Atrium TM, Hudson, NH, USA) in complex
123

Fig. 1 a, b Intraoperative
photographs. The axillary artery
is dissected free of the
surrounding tissues. The patient
is anticoagulated with heparin
(single bolus of 60–70 units/
kg), and the artery is controlled
with vessel loops (a). A
transverse incision (black
hollow arrow) is performed at
the level of the umbilicus in the
abdominal right lower quadrant
to expose the right common
iliac vein (b)
hemodialysis access patients. Demographics, comorbidi-
ties, complications, reinterventions, and patency rates were
assessed, and these variables were reported according to
the Society of Vascular Surgery (SVS) standards [8].
Descriptive statistics for demographic and clinical vari-
ables were conducted. All tests were performed using the
statistical software STATA version 14.0.
Preoperative assessment and surgical technique
We completed a thorough physical examination, including
a peripheral vascular assessment and preoperative ultra-
sound (US) vein mapping. All patients had prolonged
history on hemodialysis (average time of 7 years), with
multiple central venous catheters. Anatomically, all had
unsuitable peripheral veins in upper extremity, multiple
central venous occlusions in the thorax, and inadequate
femoral veins. Venous outflow was assessed by digital
venography, subclavian, and brachiocephalic veins, and the
iliocaval systems were imaged and the adequate outflow
vein was selected. The inflow from the axillary artery was
selected based on previous access surgeries and optimal
pulse examination.
Operative technique
All cases were done under general anesthesia and opera-
tions performed by one staff surgeon (CH). A 50 cm
conduit and 6 mm in diameter EC graft are selected, and
the length is adjusted according to each patient. The patient
is placed in supine position with the elevation of the ipsi-
lateral flank to the venous anastomosis using a soft roll.
Wide prepping and draping is critical to allow the exposure
of the clavicle, lower neck, anterior chest, abdomen, and
both groins. The first portion of the axillary artery, between
123
J Artif Organs
the clavicle and pectoralis minor muscle, is exposed using
a transverse incision of approximately 6 cm in length. The
deep fascia is incised parallel to the incision, and the
pectoralis major fibers are split, exposing the fat containing
the axillary artery, vein, and vessel branches. The artery is
dissected free of the surrounding tissues, and small bran-
ches are ligated (Fig. 1a). The patient is anticoagulated
with heparin (single bolus of 60–70 units per kilogram
[kg]), and the artery is controlled with vascular clamps. An
arteriotomy is created, the axillary end of the graft is
spatulated, and the anastomosis is performed with
6–0 prolene. Subsequently, the graft is tunneled in the
anterior chest wall medial to the nipple with a tubular
tunneler and extended further to the upper abdomen, taking
care not to allow it to twist during the passage. A transverse
incision is performed at the level of the umbilicus in the
abdominal lower quadrants, and the laterality depends on
the adequate venous outflow in the iliocaval system
(Fig. 1b); the rectus muscle is retracted medially and the
common iliac vein is exposed through a retroperitoneal
approach, the graft is tunneled and enters the abdomen
through the fascia and fibers of the oblique muscle
approximately 5–10 cm above the incision, and an end to
side anastomosis is completed using prolene 6–0. A
handheld Doppler probe is utilized to assess flow and once
satisfied, and the wounds are closed by layers (Fig. 2a, b).
Results
In 8 patients (75 % men; mean age 37 ± SD 10 years), a
total eight 6 mm (mm) grafts were implanted. Technical
success was 100 %, and all grafts were patent immediately
after placement. Patient’s demographics and characteristics
are listed in Table 1. The average number of previous
central venous catheters (CVC) was 4. There was no pro-
cedure-related morbidity or mortality, and 7 patients
J Artif Organs

Fig. 2 a, b Early cannulation

graft is tunneled in the anterior
chest and abdominal wall, the
devices enters the abdomen
through the fascia and the fibers
of the oblique muscle (black
hollow arrow), and an end to
side anastomosis is performed
in the common iliac vein (a).
Schematic drawing representing
the configuration and trajectory
of the axillo-iliac arteriovenous
graft (AVG) (b)

Table 1 Patient’s demographic comorbidities limb edema secondary to venous hypertension were not
observed. The remaining devices are still functioning.
Number (%)

Patient’s age (years, standard deviation) 37 ± 10 –

Male gender 6 75 Discussion
Type 2 diabetes mellitus 1 12
Arterial hypertension 5 62 Autologous arteriovenous fistulas (AVF) constitute the first
Coronary artery disease 0 0 option for hemodialysis access creation as recommended
Dyslipidemia 0 0 by the ‘‘Fistula First’’ initiative and the update of the
Thrombophilia 0 0 Kidney Disease Outcomes Quality Initiative (KDOQI)
guidelines [9]. As a result of the longer survival of the
ESRD population and the prolonged waiting time period
Table 2 Axillo-iliac arteriovenous graft complications for renal transplant, vascular access must be available for
Number (%) longer periods of time; unfortunately, no all patients have
adequate veins for autologous access [10]. Surgical tech-
Hematoma 3 37
niques in the setting of exhausted peripheral sites and
Infection 2 25
occlusion of the central veins have been reported, and they
Thrombosis 5 62
typically require innovative approaches to find venous
Steal Syndrome 0 0 outflow for the creation of hemodialysis circuits [2, 3, 11].
Venous Hypertension 0 0 These complex procedures have demonstrated benefit to
Pseudoaneurysm formation 0 0 patients by decreasing the dependence on central venous
High-output heart failure 0 0 catheter (CVC)-based hemodialysis and morbidity related
to CVC infections, therefore, improving quality of life,
particularly in those who are not candidates for kidney
(88 %) successfully used the implanted device for transplant [10]. The Flixene graft was designed to provide
hemodialysis within 72 h and one (12 %) in 96 h. Com- an alternative to CVC for patients requiring urgent
plications rates are listed in Table 2, and during the mean hemodialysis by allowing the early cannulation. Primary
follow-up period of 6 months, five (62 %) patients required and secondary patency rates at 6 months were equivalent to
graft thrombectomy, one (12 %) of them balloon angio- other data reported on PTFE grafts [12]. Since the intro-
plasty at the vein anastomosis because of stenosis, and 2 duction of this device, the greatest benefit of this graft has
(25 %) grafts were removed secondary to infection. Com- been the avoidance and shorter duration of catheter use
plications as high-output heart failure, steal syndrome, and [12]. Our initial experience using this graft for complex

123

access creation includes one case of axillary to the right
atrial graft which has remained functional for 36 months,
and 10 cases using axillo-femoral AVGs. We encountered
a significant infection rate at the groin (50 %) with the
axillo-femoral configuration; based on this experience [13],
in an attempt to avoid hostile groins or because of inade-
quate femoral veins, we create the anastomosis at the
common iliac veins.
As aforementioned, the axillary-iliac AVG was first
reported by Cimochowski; interestingly, during the follow-
up of 12 months, the authors did not report episodes of
thrombosis or infections in their series. Among the com-
plications that occurred, there were three cases of venous
hypertension secondary to restenosis manifested with limb
edema, and two patients were treated with femoral venous
graft from the left femoral vein at 6 and 8 months,
respectively. At the end of the 12 month period, six of
eight patients have a functioning graft and two patients
died during the follow-up.
We made a few modifications on this technique; the first
was tunneling the graft medially to the nipple to increase the
access sites and facilitates cannulation in the anterior chest,
which is the area that we recommend as the primary site.
The two cases of infection occurred when the graft was
cannulated at its abdominal portion. The second modifica-
tion that we implemented was the anastomosis at the com-
mon iliac vein with the aim to provide a better venous
outflow. Critical technical points include the avoidance of
crossing points and angling of the device; in addition, the
retroperitoneal exposure implies a more complex procedure
and difficulties for future kidney transplantation, and this
needs to be carefully assessed and considered. None of the
patients studied in this series were candidate for peritoneal
dialysis or had failed kidney transplant.
Other’s authors that experience with this access con-
figuration include Jakimowicz et al. [7]. They reported the
largest series that included 19 patients, 5 of them required
thrombectomy, one surgical revision and graft extension to
the inferior vena cava was necessary in one patient, and 1
graft was removed at 14 months secondary to infection.
During the follow-up, one death occurred secondary to
pneumonia. In 2006, Hamish’ group [5] reported their
experience with axillo-iliac bypasses associated with dial-
ysis access, and in three patients, a venous axillo-iliac
bypass was used to treat symptoms of venous hypertension
and five axillo-iliac AVGs for access creation. The mean
follow-up was 13.2 months with secondary patency rate of
80 % at 6 months.
When all alternatives in upper extremities have been
exhausted, lower limb access represents the following
option as recommended by the clinical practice guidelines,
and we routinely follow the access algorithm [14]. Femoral
AVGs either in the upper thigh or mid-thigh or the creation
123
J Artif Organs
of autologous access transposing the femoral vein are
available. In 2009, Antoniou et al. completed a systematic
review [15], and the authors concluded that lower
extremity accesses overall had acceptable outcomes in
terms of patency, with femoral vein transposition (FVT)
arteriovenous (AV) fistula having better results than grafts.
Autologous access was associated with less rate of infec-
tion compared to synthetic grafts (1.61 vs 18 %), however,
at the expense of increased ischemic complications as a
result steal syndrome (20.97 vs 7.18 %). There are no
questions regarding the superiority of durability of FVT
AV fistula over AVGs that were successfully created, but
considering the magnitude surgery, patients require careful
selection due to the risk of ischemic complications asso-
ciated with steal syndrome and development of congestive
heart failure [16]. In the present series with axillo-iliac
AVG, we did not observe upper extremity steal syndrome
or limb edema associated with venous hypertension com-
plications observed in lower extremity AVG, neither high-
output heart failure occurred.
Conclusion
Construction of axillo-iliac AVG in the setting of exhaus-
ted superficial veins, occluded intrathoracic central veins,
and inadequates femoral veins or hostile groins that rep-
resent a viable arteriovenous access alternative. The EC
graft access allows avoidance of central venous catheters
while providing reliable AV access. Further experience is
necessary to evaluate if this graft configuration has clinical
advantages compared to thigh AVGs.
Compliance with ethical standards
Conflict of interest The authors have no conflict of interest.
References
1. Chemla ES, Morsy M, Anderson L, Makanjuola D. Complex
bypasses and fistulas for difficult hemodialysis access: a
prospective, single-center experience. Semin Dial. 2006;19:
246–50.
2. Anaya-Ayala JE, Bellows PH, Ismail N, Cheema ZF, Naoum JJ,
Bismuth J, Lumsden AB, Reardon MJ, Davies MG, Peden EK.
Surgical management of hemodialysis-related central venous
occlusive disease: a treatment algorithm. Ann Vasc Surg.
2011;25:108–19. doi:10.1016/j.avsg.2010.11.002.
3. Zamani N, Anaya-Ayala JE, Ismail N, Peden EK. Left common
iliac artery to inferior vena cava abdominal wall arteriovenous
graft for hemodialysis access. J Vasc Surg. 2012;56:489–91.
doi:10.1016/j.jvs.2012.01.062 epub 10 Apr 2012.
4. Cimochowski GE, Harter HR, Rutherford WE, Blondin J, Sartain
JA. Axillary artery to iliac vein vascular access using an exter-
nally supported prosthetic graft. A new procedure for the recal-
citrant secondary access patient. ASAIO Trans. 1987;33:123–8.
J Artif Organs
5. Hamish M, Shalhoub J, Rodd CD, Davies AH. Axillo-iliac con-
duit for haemodialysis vascular access. Eur J Vasc Endovasc
Surg. 2006;31:530–4 epub 19 Jan 2006.
6. Evans DC, Upton EC, Lawson JH. Axillary to common iliac
arteriovenous graft for hemodialysis access: Case report and
review of exotic axillary-based grafts. J Vasc Access.
2005;6:192–5.
7. Jakimowicz T, Galazka Z, Grochowiecki T, Nazarewski S,
Szmidt J. Vascular access for haemodialysis in patients with
central vein thrombosis. Eur J Vasc Endovasc Surg.
2011;42:842–9. doi:10.1016/j.ejvs.2011.07.022 epub 17 Aug
2011.
8. Sidawy AN, Gray R, Besarab A, Henry M, Ascher E, Silva Jr M,
Miller A, Scher L, Trerotola S, Gregory RT, Rutherford RB, Kent
KC. Recommended standards for reports dealing with arteriove-
nous hemodialysis accesses. J Vasc Surg. 2002;35:603–10
review.
9. Navuluri R, Regalado S. The KDOQI 2006 vascular access
update and fistula first program synopsis. Semin Interv Radiol.
2009;26:122–4. doi:10.1055/s-0029-1222455.
10. Hakaim AG, Scott TE. Durability of early prosthetic dialysis graft
cannulation: results of a prospective, nonrandomized clinical
trial. J Vasc Surg. 1997;25:1002–5 discussion 1005–6.
11. Bellows PH, Anaya-Ayala JE, Cheema ZF, Davies MG, Lumsden
AB, Peden EK. Combined femoral vein transposition and iliac
vein to suprarenal vena cava bypass as a last resort dialysis
access. Ann Vasc Surg. 2011;25:264-e5–8. doi:10.1016/j.avsg.
2010.03.029 epub 30 Aug 2010.
12. Schild AF, Schuman ES, Noicely K, Ka ufman J, Gillaspie E,
Fuller J, Collier P, Ronfeld A, Nair R. Early cannulation pros-
thetic graft (FlixeneTM) for arteriovenous access. J Vasc Access.
2011;12:248–52.
13. Hinojosa CA, Soto-Solis S, Olivarez-Cruz S, Laparra-Escareno
H, Gomez-Arcive Z, Anaya-Ayala JE. Early cannulation graft
Flixene for conventional and complex hemodialysis access cre-
ation. J Vasc Access (in press)
14. Sidawy AN, Spergel LM, Besarab A, Allon M, Jennings WC,
Padberg FT Jr, Murad MH, Montori VM, O’Hare AM, Calligaro
KD, Macsata RA, Lumsden AB, Ascher E. Society for Vascular
Surgery. The Society for Vascular Surgery: clinical practice
guidelines for the surgical placement and maintenance of arteri-
ovenous hemodialysis access. J Vasc Surg. 2008;48:2S–25S.
doi:10.1016/j.jvs.2008.08.042 review.
15. Antoniou GA, Lazarides MK, Georgiadis GS, Sfyroeras GS,
Nikolopoulos ES, Giannoukas AD. Lower-extremity arteriove-
nous access for haemodialysis: a systematic review. Eur J Vasc
Endovasc Surg. 2009;38:365–72. doi:10.1016/j.ejvs.2009.06.003
epub 10 Jul 2009, review.
16. Bourquelot P, Rawa M, Van Laere O, Franco G. Long-term
results of femoral vein transposition for autogenous arteriovenous
hemodialysis access. J Vasc Surg. 2012;56:440–5. doi:10.1016/j.
jvs.2012.01.068 epub 9 May 2012.
123