- Email: [email protected]
Outcomes of a Polytetrafluoroethylene Hybrid Vascular Graft with Preloaded Nitinol Stent at the Venous Outflow for Dialysis Vascular Access
Clinical Research Outcomes of a Polytetrafluoroethylene Hybrid Vascular Graft with Preloaded Nitinol Stent at the Venous Outflow for Dialysis Vascular Access Peiman Habibollahi,1,2,3 Mark P. Mantell,4,5,6 Trish Rosenberry,7 David B. Leeser,8 and Timothy W.I. Clark,1,2,3 Philadelphia, Pennsylvania, Queenstown, Maryland, and Greenville, North Carolina
Background: To evaluate outcomes and patency of arteriovenous grafts (AVGs) created using Gore hybrid vascular grafts in hemodialysis patients with limited venous outflow or challenging anatomy. Materials and methods: A retrospective review was performed in two academic centers of all patients between July 2013 and December 2016 who underwent surgical AVG creation using a Gore hybrid vascular graft in a brachial artery to axillary configuration. Patient characteristics and comorbidities as well as graft patency, function, and subsequent need for percutaneous interventions were recorded. Results: Forty-six patients including 30 females (65.2%) and 16 males (34.8%) with a mean age of 63 ± 13 years were identified. The most common indications for a hybrid vascular graft were limited surgical accessibility and/or revision of existing AVG due to severe stenotic lesions at the venous outflow in 33 patients (72%). One-year primary unassisted and assisted patency rates were 44 ± 8% and 54 ± 8%, respectively, compared with 1-year secondary patency rate of 66 ± 8%. The rate of percutaneous interventions to maintain graft function and patency was approximately one intervention per graft per year. Conclusions: Access created with the hybrid vascular graft in a brachial-axillary (brachial artery to axillary vein) configuration is an acceptable option for patients with limited venous outflow reserve and challenging anatomy. Twelve-month primary and secondary patency rates and need for percutaneous interventions were comparable to traditional AVGs.
INTRODUCTION Patients with end-stage renal disease (ESRD) require durable vascular access allowing repetitive Conflict of Interest: None. 1 Division of Interventional Radiology, Department of Radiology, Philadelphia, PA. 2 Penn Presbyterian Medical Center, Philadelphia, PA. 3
University of Pennsylvania, Philadelphia, PA.
4
Division of Vascular Surgery, Department of Surgery, Philadelphia, PA. 5
Penn Presbyterian Medical Center, Philadelphia, PA.
6
University of Pennsylvania, Philadelphia, PA.
7
University of Maryland Shore Regional Health, Queenstown, MD.
cannulation for hemodialysis. Owing to lower morbidity, mortality, longer patency, and lower need for interventions to maintain patency,
8 Division of Transplant Surgery, Department of Surgery, East Carolina University, Greenville, NC.
Correspondence to: Timothy W.I. Clark, MD, Associate Professor of Radiology, University of Pennsylvania Perelman School of Medicine, Penn Presbyterian Medical Center, 51 N 39th Street, Philadelphia, PA 19104; E-mail: [email protected] Ann Vasc Surg 2018; -: 1–6 https://doi.org/10.1016/j.avsg.2018.06.029 Ó 2018 Elsevier Inc. All rights reserved. Manuscript received: January 28, 2018; manuscript accepted: June 28, 2018; published online: - - -
1
2 Habibollahi et al.
arteriovenous fistulas (AVFs) using autologous conduit are recommended over AV grafts (AVG).1,2 However, increased life-expectancy of hemodialysisdependent patients results in depletion of native venous anatomy, such that many patients require prosthetic AVGs as the second preferred method of AV access creation.3 The risk of developing stenotic lesions in the outflow vein is high for AVGs; 70e80% of grafts develop stenosis of the outflow vein or thrombosis within a year from creation.4,5 Most AVG stenoses occur at the site of venous anastomosis,5 which can also make creation of future access more challenging. Venous outflow stenosis in AVGs initiates from endothelial injury from shear stress and turbulent flow6,7 resulting in neointimal hyperplasia.8,9 The Gore hybrid device consists of a polytetrafluoroethylene graft with a nitinol stent incorporated into the venous outflow end. The stent-graft outflow of the device can be deployed within the outflow vein in a sutureless, straight configuration, with the goal of decreasing turbulent flow as a contributing cause of neointimal hyperplasia. Limited outcomes data exist as to the extent of these potential benefits conferring a decrease in access-related complications.10,11 We performed a retrospective study in two tertiary referral centers to evaluate the clinical outcomes, patency, and complications of this hybrid vascular graft for hemodialysis access.
MATERIALS AND METHODS A retrospective multicenter study was performed between July 2013 and December 2016 at two academic medical centers. Institutional review board approved the study, and waiver of informed consent was obtained given the retrospective nature of the study. All patients undergoing upper extremity hemodialysis vascular access placement during the study period using the Gore hybrid vascular graft were included. Electronic medical records (EMRs) for each patient were evaluated, and patient demographics including age, gender, body mass index, associated comorbid conditions (diabetes, hypertension, hyperlipidemia and so forth), and any documented form of prior vascular access were recorded. All patients underwent preoperative evaluation in the office. Preoperative evaluation typically included a detailed medical and surgical history and a thorough physical examination including a complete vascular examination. Appropriate imaging studies such as sonographic vein mapping and, if required, upper extremity venograms were performed.
Annals of Vascular Surgery
Access Creation A single vascular surgeon in each center performed all access-creation procedures. The axillary vein was identified using ultrasound and accessed using a micropuncture kit through a small skin incision. After insertion of a 5F catheter over a 0.018-inch guidewire, diagnostic venography was performed to confirm patency of the axillary vein. A transverse incision was made just above the antecubital fossa over the brachial artery, and dissection was carried down through the subcutaneous tissues until the brachial artery was exposed. A peelaway sheath with appropriate size for the vein and graft was inserted over a 0.035-inch guidewire into the axillary vein. An appropriate weight-based dose of heparin was given. The Gore hybrid graft was inserted through the peelaway sheath and deployed under fluoroscopic guidance. The graft was then angioplastied using a balloon to the appropriate size (7e9 mm based on size of the nitinol stent in the graft). According to individual surgeon preference, the Gore graft was either then passed through a subcutaneous tunnel to the site of exposed brachial artery and anastomosed end-to-side to the artery (18 patients, nontapered graft group) or cut and anastomosed end-to-end to a tapered PTFE graft which had been anastomosed end-toside from the brachial artery and tunneled to the site of the Gore hybrid graft insertion (28 patients, tapered graft group). The EMR and the local interventional radiology database (HI-IQ, Lincoln, RI) were reviewed for immediate perioperative complications, graft patency, and short- and long-term complications such as stenosis, thrombosis, and infection. The need for subsequent endovascular and/or surgical revision to maintain access patency was recorded. As per recommendations of the Society for Vascular Surgery,12 primary patency and primary assisted patency were defined as the time interval from access creation to any intervention to reinstitute or improve flow and graft thrombosis. Secondary patency was defined as the time interval between access creation and access abandonment or last known follow-up, including any additional thrombectomy procedures to maintain patency. Numerical variables are reported as mean ± standard deviation or median. Categorical variables were reported as frequency (percentage). Univariate analysis was also performed to assess for predictors of patency loss. Statistical analysis was performed using SPSS software (V.20; IBM Corp, Armonk, NY).
Volume
-, -
2018
RESULTS Between July 2013 and December 2016, 47 patients underwent access placement using a hybrid vascular graft in both the institutions. One patient from tapered graft group was excluded secondary to loss to follow-up after surgery, resulting in 28 patients in tapered graft group and 18 patients in nontapered graft group. The size of the nitinol-reinforced segment of the most commonly used hybrid graft was 8 mm 5 cm. The sizes of the nitinolreinforced section of the grafts are listed in Table I. In 2 patients, due to extravasations seen during intraoperative venography after graft deployment, additional stent grafts were deployed central to the nitinol-reinforced section of the graft (8 mm 6 cm and 10 mm 5 cm). Table II lists baseline demographics and patient characteristics; overall, limited surgical accessibility was the most common reason to choose the hybrid graft. Eight patients received the hybrid graft as their first access for this reason. The mean follow-up duration for the study population was 358 days. The mean follow-up was longer for patients in the tapered graft group, but the difference was not statistically significant (299 days for tapered graft group versus 450 days for in nontapered graft group, P ¼ 0.14). Time to access cannulation was available in 17 patients with a mean of 25 ± 14 days from access placement (range: 9e61 days). Mean primary unassisted patency duration for the entire study population was 492 ± 84 with a median of 196 days compared with mean primary assisted patency duration of 512 ± 85 days with a median of 427 days. Mean duration of secondary patency was 780 ± 87 days with a median of 1155 days. Figure 1 shows one-year patency Kaplan-Meier curves for the study population. Primary unassisted patency at the end of one-year follow-up was 44 ± 8% compared with 54 ± 8% primary assisted patency and 66 ± 8% secondary patency (Fig. 1). Graft patency (primary unassisted, primary assisted, and secondary) was compared within study population based on using tapered graft for creating graft to artery anastomosis. There was no statistically significant difference in graft patency in patients who had hybrid vascular graft anastomosed to a tapered graft compared with patents with direct end-to-side anastomosis of the hybrid vascular graft to brachial artery. For example, mean secondary patency was 781 ± 110 versus 742 ± 139 for patients in tapered graft group versus patients in nontapered graft group, respectively (P ¼ 0.65). Univariate analysis was performed using variables listed in Table II to allow identification of pa-
Hybrid graftestent graft for dialysis access 3
Table I. Size of the Nitinol-reinforced section of the hybrid grafts used in the study Dimension of stent graft component
6 7 7 8 9 9
mm mm mm mm mm mm
5 cm 5 cm 10 cm 5 cm 5 cm 10 cm
Center A (n ¼ 28)
Center B (n ¼ 18)
Overall (n ¼ 46)
0 10 1 2 12 3
5 (28%) 0 0 13 (72%) 0 0
5 10 1 15 12 3
(36%) (4%) (7%) (43%) (11%)
(11%) (22%) (2%) (33%) (26%) (7%)
tient factors associated with loss of patency during study period. No statistically significant predictor of patency loss was identified in this model (data not shown). During the study period, 45 percutaneous interventions were performed for a sum follow-up period of 16,640 days (1 intervention per 369.8 days per graft or 0.99 interventions per graft per year). Type of interventions included 26 thrombectomies and 19 fistulograms. During one thrombectomy procedure, a stent graft was placed inside the graft due to a long-segment stenosis related to repetitive cannulation. Four fistulograms out of 19 were negative and did not show any treatable lesions. Twelve fistulogram procedures involved balloon angioplasty to treat intragraft and/or perigraft stenoses; 3 stent grafts were also deployed during the remainder of fistulograms in 3 different patients for treating venous outflow lesions refractory to angioplasty.
DISCUSSION The optimal vascular access for hemodialysis in ESRD patients should allow repetitive puncture and access to the circulation with reasonable durability and minimal complications. Currently, surgically placed AVFs and AVGs are two of the most commonly used form of access in this patient population. AVFs are preferred over AVGs because of their durability, lower patient morbidity, mortality, and overall need for interventions to maintain patency.1 However, after exhaustion of native veins, most ESRD patients will require some form of prosthetic AVG placed.3 Beyond the immediate postoperative period, the most common cause of graft dysfunction and failure is the development of vascular stenotic lesions. Stenotic lesions are usually focal (up to 70%) and most commonly affect the venous outflow (up to 80%) with a high predilection to the venous anastomosis.5,12e14 Refinements in graft technologies to
4 Habibollahi et al.
Annals of Vascular Surgery
Table II. Demographic and clinical characteristics of the patients included in the study Variable a
Age (mean) Gender (female) BMIa Indication Limited surgical accessibility Obesity Revision of existing access Prior access history (multiple) Comorbidities Obesity DM HTN CHF Factor V Leiden
Tapered graft (n ¼ 28)
Nontapered graft (n ¼ 18)
Overall (n ¼ 46)
61 ± 13 19 (68%) 31 ± 7.9
66 ± 12 11 (61%) 35 ± 9.9
62 ± 12 30 (65%) 33 ± 8.8
21 5 2 25
(75%) (18%) (7%) (89%)
3 8 7 13
(17%) (44%) (39%) (72%)
24 13 9 38
(52%) (28%) (20%) (83%)
14 16 22 6 0
(50%) (57%) (79%) (21%) (0%)
9 14 12 3 1
(50%) (78%) (67%) (17%) (6%)
23 30 34 9 1
(50%) (65%) (74%) (20%) (2%)
BMI, body mass index; SD, standard deviation; HTN, hypertension; CHF, congestive heart failure; DM, diabetes mellitus. a Data are presented as mean ± SD.
Fig. 1. Kaplan-Meier plot showing survival of Gore hybrid vascular graft categorized based on primary unassisted, primary assisted, and secondary patencies.
Volume
-, -
2018
improve patient outcomes over the last decade have included efforts to reduce AVG thrombogenicity using covalently bound heparin,15 early cannulation grafts to reduce the need for dialysis catheters,16,17 and stent-graft outflow devices to reduce the development of venous anastomotic stenoses and allow placement of grafts in patients with challenging anatomy, such as the Gore hybrid graft. In the present study, we retrospectively reviewed 46 patients who underwent upper arm AVG creation using the Gore hybrid vascular graft in 2 different centers. Observed 1-year primary unassisted and assisted patency were 44 ± 8% and 54 ± 8%, respectively, with a secondary patency of 66 ± 8%. These findings were comparable to reported 1-year functional primary and secondary patency rates of 66 ± 10% and 79 ± 9%, respectively, in 32 patients by Benedetto et al.10 but markedly higher than 1-year primary and secondary patency rates of 12.5% and 25%, respectively, reported by Murga et al.14 It is worth noting that the experience of Murga et al. was limited to 8 patients and that grafts placed in their study were divided between 2 configurations of either an axilloaxillary loop graft or a brachial-axillary configuration, and also in the study by Benedetto et al.,10 different configurations were used to create the access (brachial-axillary in 22 [69%], brachialbasilic in 5 [16%], brachial-antecubital loop in 3 [9%], and axilloaxillary loop graft in 1 [3%]). In the present study, only a brachial artery to axillary vein configuration was used. Our patency rates are also similar to primary and secondary patencies of conventional PTFE AVGs at one year ranging from 40 to 60% and 50e70%, respectively,10,18,19 noting that the majority of the patients included in the present study were chosen to undergo revision of an existing graft with severe venous outflow stenosis and/or morbid obesity precluding conventional AVG placement. Another important study evaluating the patency of Gore hybrid vascular graft in comparison to PTFE grafts in patients with disadvantaged anatomy by AnayaAyala et al.13 showed one-year primary, assisted primary, and secondary patency of 66%, 69%, and 69%, respectively, compared with 45%, 50%, and 66%, respectively, for the access created using PTFE. It is also worth noting that the majority of the accesses were created in the brachial artery axillary vein configuration in this study, which is similar to the present study. Venous outflow stenosis in AVGs is thought to initiate from endothelial injury secondary to shear stress generated by turbulent flow6,7 and differences in compliance between native vein and synthetic
Hybrid graftestent graft for dialysis access 5
graft.20 These factors lead to upregulation of adhesion molecules on the endothelial surface, which initiates a cascade of events including leukocyte adherence, leukocyte migration, and smooth muscle cell hyperplasia. These events result in fibromuscular hyperplasia.21e23 One of the important factors contributing to shear stress is the turbulent flow resulting from the connection between high- and low-pressure systems at the site of an end-to-side anastomosis at the venous outflow in traditional AVGs, which has been shown previously to enhance neointimal hyperplasia.8,9 Theoretically, the sutureless and straight configuration of the venous anastomosis created by the Gore hybrid graft has the potential to decrease shear stress, turbulent flow, and subsequent intimal injury resulting in a reduction in venous outflow stenosis. In the present study, the need for percutaneous interventions to maintain AVG patency approximated 1 intervention per graft per year, divided between thrombectomies (58%) and angioplasty/stenting (42%). Previous studies have shown that traditional AVGs require 1.05 percutaneous interventions per graft per year and 0.17 surgical revisions per graft per year to maintain function and patency,24 which is comparable to our findings. The present study consisted of patients with challenging anatomy and poor venous outflow options. This may indicate the use of the Gore hybrid graft in this particular patient population. The present study has multiple limitations. The patient population was heterogeneous, and no predefined selection process was involved due to the retrospective nature of the study. The follow-up period was variable. Differences in patient selection and surgical technique would also be inherent to two high-volume access centers with differing geographic patterns of patient referral. In conclusion, AVGs created with the Gore hybrid vascular graft in a brachial-axillary configuration are an alternative for traditional AVGs in patients with limited venous outflow with comparable 1year primary and secondary patencies and need for percutaneous interventions to maintain function and patency. REFERENCES 1. Hemodialysis Adequacy Work G. Clinical practice guidelines for hemodialysis adequacy, update 2006. Am J Kidney Dis 2006;48(Suppl 1):S2e90. 2. Polkinghorne KR, McDonald SP, Atkins RC, et al. Vascular access and all-cause mortality: a propensity score analysis. J Am Soc Nephrol 2004;15:477e86. 3. Akoh JA. Prosthetic arteriovenous grafts for hemodialysis. J Vasc Access 2009;10:137e47.
6 Habibollahi et al.
4. Dixon BS, Beck GJ, Vazquez MA, et al. Effect of dipyridamole plus aspirin on hemodialysis graft patency. N Engl J Med 2009;360:2191e201. 5. Maya ID, Oser R, Saddekni S, et al. Vascular access stenosis: comparison of arteriovenous grafts and fistulas. Am J Kidney Dis 2004;44:859e65. 6. Hsieh HJ, Li NQ, Frangos JA. Shear stress increases endothelial platelet-derived growth factor mRNA levels. Am J Physiol 1991;260(2 Pt 2):H642e6. 7. Sterpetti AV, Cucina A, Santoro L, et al. Modulation of arterial smooth muscle cell growth by haemodynamic forces. Eur J Vasc Surg 1992;6:16e20. 8. Heise M, Schmidt S, Kruger U, et al. Local haemodynamics and shear stress in cuffed and straight PTFE-venous anastomoses: an in-vitro comparison using particle image velocimetry. Eur J Vasc Endovasc Surg 2003;26:367e73. 9. Krueger U, Huhle A, Krys K, et al. Effect of tapered grafts on hemodynamics and flow rate in dialysis access grafts. Artif Organs 2004;28:623e8. 10. Benedetto F, Spinelli D, Pipito N, et al. Initial clinical experience with a polytetrafluoroethylene vascular dialysis graft reinforced with nitinol at the venous end. J Vasc Surg 2017;65:142e50. 11. Teruya TH, Schaeffer D, Abou-Zamzam AM, et al. Arteriovenous graft with outflow in the proximal axillary vein. Ann Vasc Surg 2009;23:95e8. 12. Sidawy AN, Gray R, Besarab A, et al. Recommended standards for reports dealing with arteriovenous hemodialysis accesses. J Vasc Surg 2002;35:603e10. 13. Anaya-Ayala JE, Davies MG, El-Sayed HF, et al. Early experience with a Novel hybrid vascular graft for hemodialysis access creation in patients with disadvantaged anatomy. J Endovasc Ther 2015;22:778e85. 14. Murga AG, Chiriano J, Kiang SC, et al. Arteriovenous hybrid graft with outflow in the proximal axillary vein. Ann Vasc Surg 2017;42:39e44.
Annals of Vascular Surgery
15. Walluscheck KP, Bierkandt S, Brandt M, et al. Infrainguinal ePTFE vascular graft with bioactive surface heparin bonding. First clinical results. J Cardiovasc Surg (torino) 2005;46:425e30. 16. Berard X, Ottaviani N, Brizzi V, et al. Use of the Flixene vascular access graft as an early cannulation solution. J Vasc Surg 2015;62:128e34. 17. Tozzi M, Franchin M, Ietto G, et al. Initial experience with the Gore(R) Acuseal graft for prosthetic vascular access. J Vasc Access 2014;15:385e90. 18. Gibson KD, Gillen DL, Caps MT, et al. Vascular access survival and incidence of revisions: a comparison of prosthetic grafts, simple autogenous fistulas, and venous transposition fistulas from the United States Renal Data System Dialysis Morbidity and Mortality Study. J Vasc Surg 2001;34: 694e700. 19. Huber TS, Carter JW, Carter RL, et al. Patency of autogenous and polytetrafluoroethylene upper extremity arteriovenous hemodialysis accesses: a systematic review. J Vasc Surg 2003;38:1005e11. 20. Hofstra L, Bergmans DC, Hoeks AP, et al. Mismatch in elastic properties around anastomoses of interposition grafts for hemodialysis access. J Am Soc Nephrol 1994;5:1243e50. 21. Rekhter M, Nicholls S, Ferguson M, et al. Cell proliferation in human arteriovenous fistulas used for hemodialysis. Arterioscler Thromb 1993;13:609e17. 22. Stracke S, Konner K, Kostlin I, et al. Increased expression of TGF-beta1 and IGF-I in inflammatory stenotic lesions of hemodialysis fistulas. Kidney Int 2002;61:1011e9. 23. Swedberg SH, Brown BG, Sigley R, et al. Intimal fibromuscular hyperplasia at the venous anastomosis of PTFE grafts in hemodialysis patients. Clinical, immunocytochemical, light and electron microscopic assessment. Circulation 1989;80:1726e36. 24. Miller PE, Carlton D, Deierhoi MH, et al. Natural history of arteriovenous grafts in hemodialysis patients. Am J Kidney Dis 2000;36:68e74.
Background: To evaluate outcomes and patency of arteriovenous grafts (AVGs) created using Gore hybrid vascular grafts in hemodialysis patients with limited venous outflow or challenging anatomy. Materials and methods: A retrospective review was performed in two academic centers of all patients between July 2013 and December 2016 who underwent surgical AVG creation using a Gore hybrid vascular graft in a brachial artery to axillary configuration. Patient characteristics and comorbidities as well as graft patency, function, and subsequent need for percutaneous interventions were recorded. Results: Forty-six patients including 30 females (65.2%) and 16 males (34.8%) with a mean age of 63 ± 13 years were identified. The most common indications for a hybrid vascular graft were limited surgical accessibility and/or revision of existing AVG due to severe stenotic lesions at the venous outflow in 33 patients (72%). One-year primary unassisted and assisted patency rates were 44 ± 8% and 54 ± 8%, respectively, compared with 1-year secondary patency rate of 66 ± 8%. The rate of percutaneous interventions to maintain graft function and patency was approximately one intervention per graft per year. Conclusions: Access created with the hybrid vascular graft in a brachial-axillary (brachial artery to axillary vein) configuration is an acceptable option for patients with limited venous outflow reserve and challenging anatomy. Twelve-month primary and secondary patency rates and need for percutaneous interventions were comparable to traditional AVGs.
INTRODUCTION Patients with end-stage renal disease (ESRD) require durable vascular access allowing repetitive Conflict of Interest: None. 1 Division of Interventional Radiology, Department of Radiology, Philadelphia, PA. 2 Penn Presbyterian Medical Center, Philadelphia, PA. 3
University of Pennsylvania, Philadelphia, PA.
4
Division of Vascular Surgery, Department of Surgery, Philadelphia, PA. 5
Penn Presbyterian Medical Center, Philadelphia, PA.
6
University of Pennsylvania, Philadelphia, PA.
7
University of Maryland Shore Regional Health, Queenstown, MD.
cannulation for hemodialysis. Owing to lower morbidity, mortality, longer patency, and lower need for interventions to maintain patency,
8 Division of Transplant Surgery, Department of Surgery, East Carolina University, Greenville, NC.
Correspondence to: Timothy W.I. Clark, MD, Associate Professor of Radiology, University of Pennsylvania Perelman School of Medicine, Penn Presbyterian Medical Center, 51 N 39th Street, Philadelphia, PA 19104; E-mail: [email protected] Ann Vasc Surg 2018; -: 1–6 https://doi.org/10.1016/j.avsg.2018.06.029 Ó 2018 Elsevier Inc. All rights reserved. Manuscript received: January 28, 2018; manuscript accepted: June 28, 2018; published online: - - -
1
2 Habibollahi et al.
arteriovenous fistulas (AVFs) using autologous conduit are recommended over AV grafts (AVG).1,2 However, increased life-expectancy of hemodialysisdependent patients results in depletion of native venous anatomy, such that many patients require prosthetic AVGs as the second preferred method of AV access creation.3 The risk of developing stenotic lesions in the outflow vein is high for AVGs; 70e80% of grafts develop stenosis of the outflow vein or thrombosis within a year from creation.4,5 Most AVG stenoses occur at the site of venous anastomosis,5 which can also make creation of future access more challenging. Venous outflow stenosis in AVGs initiates from endothelial injury from shear stress and turbulent flow6,7 resulting in neointimal hyperplasia.8,9 The Gore hybrid device consists of a polytetrafluoroethylene graft with a nitinol stent incorporated into the venous outflow end. The stent-graft outflow of the device can be deployed within the outflow vein in a sutureless, straight configuration, with the goal of decreasing turbulent flow as a contributing cause of neointimal hyperplasia. Limited outcomes data exist as to the extent of these potential benefits conferring a decrease in access-related complications.10,11 We performed a retrospective study in two tertiary referral centers to evaluate the clinical outcomes, patency, and complications of this hybrid vascular graft for hemodialysis access.
MATERIALS AND METHODS A retrospective multicenter study was performed between July 2013 and December 2016 at two academic medical centers. Institutional review board approved the study, and waiver of informed consent was obtained given the retrospective nature of the study. All patients undergoing upper extremity hemodialysis vascular access placement during the study period using the Gore hybrid vascular graft were included. Electronic medical records (EMRs) for each patient were evaluated, and patient demographics including age, gender, body mass index, associated comorbid conditions (diabetes, hypertension, hyperlipidemia and so forth), and any documented form of prior vascular access were recorded. All patients underwent preoperative evaluation in the office. Preoperative evaluation typically included a detailed medical and surgical history and a thorough physical examination including a complete vascular examination. Appropriate imaging studies such as sonographic vein mapping and, if required, upper extremity venograms were performed.
Annals of Vascular Surgery
Access Creation A single vascular surgeon in each center performed all access-creation procedures. The axillary vein was identified using ultrasound and accessed using a micropuncture kit through a small skin incision. After insertion of a 5F catheter over a 0.018-inch guidewire, diagnostic venography was performed to confirm patency of the axillary vein. A transverse incision was made just above the antecubital fossa over the brachial artery, and dissection was carried down through the subcutaneous tissues until the brachial artery was exposed. A peelaway sheath with appropriate size for the vein and graft was inserted over a 0.035-inch guidewire into the axillary vein. An appropriate weight-based dose of heparin was given. The Gore hybrid graft was inserted through the peelaway sheath and deployed under fluoroscopic guidance. The graft was then angioplastied using a balloon to the appropriate size (7e9 mm based on size of the nitinol stent in the graft). According to individual surgeon preference, the Gore graft was either then passed through a subcutaneous tunnel to the site of exposed brachial artery and anastomosed end-to-side to the artery (18 patients, nontapered graft group) or cut and anastomosed end-to-end to a tapered PTFE graft which had been anastomosed end-toside from the brachial artery and tunneled to the site of the Gore hybrid graft insertion (28 patients, tapered graft group). The EMR and the local interventional radiology database (HI-IQ, Lincoln, RI) were reviewed for immediate perioperative complications, graft patency, and short- and long-term complications such as stenosis, thrombosis, and infection. The need for subsequent endovascular and/or surgical revision to maintain access patency was recorded. As per recommendations of the Society for Vascular Surgery,12 primary patency and primary assisted patency were defined as the time interval from access creation to any intervention to reinstitute or improve flow and graft thrombosis. Secondary patency was defined as the time interval between access creation and access abandonment or last known follow-up, including any additional thrombectomy procedures to maintain patency. Numerical variables are reported as mean ± standard deviation or median. Categorical variables were reported as frequency (percentage). Univariate analysis was also performed to assess for predictors of patency loss. Statistical analysis was performed using SPSS software (V.20; IBM Corp, Armonk, NY).
Volume
-, -
2018
RESULTS Between July 2013 and December 2016, 47 patients underwent access placement using a hybrid vascular graft in both the institutions. One patient from tapered graft group was excluded secondary to loss to follow-up after surgery, resulting in 28 patients in tapered graft group and 18 patients in nontapered graft group. The size of the nitinol-reinforced segment of the most commonly used hybrid graft was 8 mm 5 cm. The sizes of the nitinolreinforced section of the grafts are listed in Table I. In 2 patients, due to extravasations seen during intraoperative venography after graft deployment, additional stent grafts were deployed central to the nitinol-reinforced section of the graft (8 mm 6 cm and 10 mm 5 cm). Table II lists baseline demographics and patient characteristics; overall, limited surgical accessibility was the most common reason to choose the hybrid graft. Eight patients received the hybrid graft as their first access for this reason. The mean follow-up duration for the study population was 358 days. The mean follow-up was longer for patients in the tapered graft group, but the difference was not statistically significant (299 days for tapered graft group versus 450 days for in nontapered graft group, P ¼ 0.14). Time to access cannulation was available in 17 patients with a mean of 25 ± 14 days from access placement (range: 9e61 days). Mean primary unassisted patency duration for the entire study population was 492 ± 84 with a median of 196 days compared with mean primary assisted patency duration of 512 ± 85 days with a median of 427 days. Mean duration of secondary patency was 780 ± 87 days with a median of 1155 days. Figure 1 shows one-year patency Kaplan-Meier curves for the study population. Primary unassisted patency at the end of one-year follow-up was 44 ± 8% compared with 54 ± 8% primary assisted patency and 66 ± 8% secondary patency (Fig. 1). Graft patency (primary unassisted, primary assisted, and secondary) was compared within study population based on using tapered graft for creating graft to artery anastomosis. There was no statistically significant difference in graft patency in patients who had hybrid vascular graft anastomosed to a tapered graft compared with patents with direct end-to-side anastomosis of the hybrid vascular graft to brachial artery. For example, mean secondary patency was 781 ± 110 versus 742 ± 139 for patients in tapered graft group versus patients in nontapered graft group, respectively (P ¼ 0.65). Univariate analysis was performed using variables listed in Table II to allow identification of pa-
Hybrid graftestent graft for dialysis access 3
Table I. Size of the Nitinol-reinforced section of the hybrid grafts used in the study Dimension of stent graft component
6 7 7 8 9 9
mm mm mm mm mm mm
5 cm 5 cm 10 cm 5 cm 5 cm 10 cm
Center A (n ¼ 28)
Center B (n ¼ 18)
Overall (n ¼ 46)
0 10 1 2 12 3
5 (28%) 0 0 13 (72%) 0 0
5 10 1 15 12 3
(36%) (4%) (7%) (43%) (11%)
(11%) (22%) (2%) (33%) (26%) (7%)
tient factors associated with loss of patency during study period. No statistically significant predictor of patency loss was identified in this model (data not shown). During the study period, 45 percutaneous interventions were performed for a sum follow-up period of 16,640 days (1 intervention per 369.8 days per graft or 0.99 interventions per graft per year). Type of interventions included 26 thrombectomies and 19 fistulograms. During one thrombectomy procedure, a stent graft was placed inside the graft due to a long-segment stenosis related to repetitive cannulation. Four fistulograms out of 19 were negative and did not show any treatable lesions. Twelve fistulogram procedures involved balloon angioplasty to treat intragraft and/or perigraft stenoses; 3 stent grafts were also deployed during the remainder of fistulograms in 3 different patients for treating venous outflow lesions refractory to angioplasty.
DISCUSSION The optimal vascular access for hemodialysis in ESRD patients should allow repetitive puncture and access to the circulation with reasonable durability and minimal complications. Currently, surgically placed AVFs and AVGs are two of the most commonly used form of access in this patient population. AVFs are preferred over AVGs because of their durability, lower patient morbidity, mortality, and overall need for interventions to maintain patency.1 However, after exhaustion of native veins, most ESRD patients will require some form of prosthetic AVG placed.3 Beyond the immediate postoperative period, the most common cause of graft dysfunction and failure is the development of vascular stenotic lesions. Stenotic lesions are usually focal (up to 70%) and most commonly affect the venous outflow (up to 80%) with a high predilection to the venous anastomosis.5,12e14 Refinements in graft technologies to
4 Habibollahi et al.
Annals of Vascular Surgery
Table II. Demographic and clinical characteristics of the patients included in the study Variable a
Age (mean) Gender (female) BMIa Indication Limited surgical accessibility Obesity Revision of existing access Prior access history (multiple) Comorbidities Obesity DM HTN CHF Factor V Leiden
Tapered graft (n ¼ 28)
Nontapered graft (n ¼ 18)
Overall (n ¼ 46)
61 ± 13 19 (68%) 31 ± 7.9
66 ± 12 11 (61%) 35 ± 9.9
62 ± 12 30 (65%) 33 ± 8.8
21 5 2 25
(75%) (18%) (7%) (89%)
3 8 7 13
(17%) (44%) (39%) (72%)
24 13 9 38
(52%) (28%) (20%) (83%)
14 16 22 6 0
(50%) (57%) (79%) (21%) (0%)
9 14 12 3 1
(50%) (78%) (67%) (17%) (6%)
23 30 34 9 1
(50%) (65%) (74%) (20%) (2%)
BMI, body mass index; SD, standard deviation; HTN, hypertension; CHF, congestive heart failure; DM, diabetes mellitus. a Data are presented as mean ± SD.
Fig. 1. Kaplan-Meier plot showing survival of Gore hybrid vascular graft categorized based on primary unassisted, primary assisted, and secondary patencies.
Volume
-, -
2018
improve patient outcomes over the last decade have included efforts to reduce AVG thrombogenicity using covalently bound heparin,15 early cannulation grafts to reduce the need for dialysis catheters,16,17 and stent-graft outflow devices to reduce the development of venous anastomotic stenoses and allow placement of grafts in patients with challenging anatomy, such as the Gore hybrid graft. In the present study, we retrospectively reviewed 46 patients who underwent upper arm AVG creation using the Gore hybrid vascular graft in 2 different centers. Observed 1-year primary unassisted and assisted patency were 44 ± 8% and 54 ± 8%, respectively, with a secondary patency of 66 ± 8%. These findings were comparable to reported 1-year functional primary and secondary patency rates of 66 ± 10% and 79 ± 9%, respectively, in 32 patients by Benedetto et al.10 but markedly higher than 1-year primary and secondary patency rates of 12.5% and 25%, respectively, reported by Murga et al.14 It is worth noting that the experience of Murga et al. was limited to 8 patients and that grafts placed in their study were divided between 2 configurations of either an axilloaxillary loop graft or a brachial-axillary configuration, and also in the study by Benedetto et al.,10 different configurations were used to create the access (brachial-axillary in 22 [69%], brachialbasilic in 5 [16%], brachial-antecubital loop in 3 [9%], and axilloaxillary loop graft in 1 [3%]). In the present study, only a brachial artery to axillary vein configuration was used. Our patency rates are also similar to primary and secondary patencies of conventional PTFE AVGs at one year ranging from 40 to 60% and 50e70%, respectively,10,18,19 noting that the majority of the patients included in the present study were chosen to undergo revision of an existing graft with severe venous outflow stenosis and/or morbid obesity precluding conventional AVG placement. Another important study evaluating the patency of Gore hybrid vascular graft in comparison to PTFE grafts in patients with disadvantaged anatomy by AnayaAyala et al.13 showed one-year primary, assisted primary, and secondary patency of 66%, 69%, and 69%, respectively, compared with 45%, 50%, and 66%, respectively, for the access created using PTFE. It is also worth noting that the majority of the accesses were created in the brachial artery axillary vein configuration in this study, which is similar to the present study. Venous outflow stenosis in AVGs is thought to initiate from endothelial injury secondary to shear stress generated by turbulent flow6,7 and differences in compliance between native vein and synthetic
Hybrid graftestent graft for dialysis access 5
graft.20 These factors lead to upregulation of adhesion molecules on the endothelial surface, which initiates a cascade of events including leukocyte adherence, leukocyte migration, and smooth muscle cell hyperplasia. These events result in fibromuscular hyperplasia.21e23 One of the important factors contributing to shear stress is the turbulent flow resulting from the connection between high- and low-pressure systems at the site of an end-to-side anastomosis at the venous outflow in traditional AVGs, which has been shown previously to enhance neointimal hyperplasia.8,9 Theoretically, the sutureless and straight configuration of the venous anastomosis created by the Gore hybrid graft has the potential to decrease shear stress, turbulent flow, and subsequent intimal injury resulting in a reduction in venous outflow stenosis. In the present study, the need for percutaneous interventions to maintain AVG patency approximated 1 intervention per graft per year, divided between thrombectomies (58%) and angioplasty/stenting (42%). Previous studies have shown that traditional AVGs require 1.05 percutaneous interventions per graft per year and 0.17 surgical revisions per graft per year to maintain function and patency,24 which is comparable to our findings. The present study consisted of patients with challenging anatomy and poor venous outflow options. This may indicate the use of the Gore hybrid graft in this particular patient population. The present study has multiple limitations. The patient population was heterogeneous, and no predefined selection process was involved due to the retrospective nature of the study. The follow-up period was variable. Differences in patient selection and surgical technique would also be inherent to two high-volume access centers with differing geographic patterns of patient referral. In conclusion, AVGs created with the Gore hybrid vascular graft in a brachial-axillary configuration are an alternative for traditional AVGs in patients with limited venous outflow with comparable 1year primary and secondary patencies and need for percutaneous interventions to maintain function and patency. REFERENCES 1. Hemodialysis Adequacy Work G. Clinical practice guidelines for hemodialysis adequacy, update 2006. Am J Kidney Dis 2006;48(Suppl 1):S2e90. 2. Polkinghorne KR, McDonald SP, Atkins RC, et al. Vascular access and all-cause mortality: a propensity score analysis. J Am Soc Nephrol 2004;15:477e86. 3. Akoh JA. Prosthetic arteriovenous grafts for hemodialysis. J Vasc Access 2009;10:137e47.
6 Habibollahi et al.
4. Dixon BS, Beck GJ, Vazquez MA, et al. Effect of dipyridamole plus aspirin on hemodialysis graft patency. N Engl J Med 2009;360:2191e201. 5. Maya ID, Oser R, Saddekni S, et al. Vascular access stenosis: comparison of arteriovenous grafts and fistulas. Am J Kidney Dis 2004;44:859e65. 6. Hsieh HJ, Li NQ, Frangos JA. Shear stress increases endothelial platelet-derived growth factor mRNA levels. Am J Physiol 1991;260(2 Pt 2):H642e6. 7. Sterpetti AV, Cucina A, Santoro L, et al. Modulation of arterial smooth muscle cell growth by haemodynamic forces. Eur J Vasc Surg 1992;6:16e20. 8. Heise M, Schmidt S, Kruger U, et al. Local haemodynamics and shear stress in cuffed and straight PTFE-venous anastomoses: an in-vitro comparison using particle image velocimetry. Eur J Vasc Endovasc Surg 2003;26:367e73. 9. Krueger U, Huhle A, Krys K, et al. Effect of tapered grafts on hemodynamics and flow rate in dialysis access grafts. Artif Organs 2004;28:623e8. 10. Benedetto F, Spinelli D, Pipito N, et al. Initial clinical experience with a polytetrafluoroethylene vascular dialysis graft reinforced with nitinol at the venous end. J Vasc Surg 2017;65:142e50. 11. Teruya TH, Schaeffer D, Abou-Zamzam AM, et al. Arteriovenous graft with outflow in the proximal axillary vein. Ann Vasc Surg 2009;23:95e8. 12. Sidawy AN, Gray R, Besarab A, et al. Recommended standards for reports dealing with arteriovenous hemodialysis accesses. J Vasc Surg 2002;35:603e10. 13. Anaya-Ayala JE, Davies MG, El-Sayed HF, et al. Early experience with a Novel hybrid vascular graft for hemodialysis access creation in patients with disadvantaged anatomy. J Endovasc Ther 2015;22:778e85. 14. Murga AG, Chiriano J, Kiang SC, et al. Arteriovenous hybrid graft with outflow in the proximal axillary vein. Ann Vasc Surg 2017;42:39e44.
Annals of Vascular Surgery
15. Walluscheck KP, Bierkandt S, Brandt M, et al. Infrainguinal ePTFE vascular graft with bioactive surface heparin bonding. First clinical results. J Cardiovasc Surg (torino) 2005;46:425e30. 16. Berard X, Ottaviani N, Brizzi V, et al. Use of the Flixene vascular access graft as an early cannulation solution. J Vasc Surg 2015;62:128e34. 17. Tozzi M, Franchin M, Ietto G, et al. Initial experience with the Gore(R) Acuseal graft for prosthetic vascular access. J Vasc Access 2014;15:385e90. 18. Gibson KD, Gillen DL, Caps MT, et al. Vascular access survival and incidence of revisions: a comparison of prosthetic grafts, simple autogenous fistulas, and venous transposition fistulas from the United States Renal Data System Dialysis Morbidity and Mortality Study. J Vasc Surg 2001;34: 694e700. 19. Huber TS, Carter JW, Carter RL, et al. Patency of autogenous and polytetrafluoroethylene upper extremity arteriovenous hemodialysis accesses: a systematic review. J Vasc Surg 2003;38:1005e11. 20. Hofstra L, Bergmans DC, Hoeks AP, et al. Mismatch in elastic properties around anastomoses of interposition grafts for hemodialysis access. J Am Soc Nephrol 1994;5:1243e50. 21. Rekhter M, Nicholls S, Ferguson M, et al. Cell proliferation in human arteriovenous fistulas used for hemodialysis. Arterioscler Thromb 1993;13:609e17. 22. Stracke S, Konner K, Kostlin I, et al. Increased expression of TGF-beta1 and IGF-I in inflammatory stenotic lesions of hemodialysis fistulas. Kidney Int 2002;61:1011e9. 23. Swedberg SH, Brown BG, Sigley R, et al. Intimal fibromuscular hyperplasia at the venous anastomosis of PTFE grafts in hemodialysis patients. Clinical, immunocytochemical, light and electron microscopic assessment. Circulation 1989;80:1726e36. 24. Miller PE, Carlton D, Deierhoi MH, et al. Natural history of arteriovenous grafts in hemodialysis patients. Am J Kidney Dis 2000;36:68e74.