A Comparison Between the HeRO Graft and Conventional Arteriovenous Grafts in Hemodialysis Patients George M. Nassar,* Marc H. Glickman,† Robert B. McLafferty,‡ J. Kevin Croston,§ Joseph I. Zarge,¶ Howard E. Katzman,** Eric K. Peden,* Jeffrey H. Lawson,‡‡ Jeffrey M. Martinez,§§ and Lisa Thackeray¶¶ *Houston Methodist Hospital, Weill Cornell University, Houston, Texas, †Sentara Heart Hospital, Norfolk, Virginia, ‡Southern Illinois University, Springfield, Illinois, §North Memorial Medical Center, Robbinsdale, Minnesota, ¶Emory St. Joseph’s Hospital, Atlanta, Georgia, **University of Miami Hospital, Miami, Florida, ‡‡Duke University Hospital, Durham, North Carolina, §§Baptist Medical Center, San Antonio, Texas, and ¶¶NAMSA, Minneapolis, Minnesota
ABSTRACT Venous stenosis and occlusion are a major cause of vascular access dysfunction and failure. The HeRO Graft bypasses occlusion and traverses stenosis with outflow directly into the central venous circulation. A randomized, multicenter study was conducted to evaluate the efficacy and safety of the HeRO Graft relative to conventional AV grafts. The design was to enroll 143 patients in a 2:1 randomization ratio between HeRO and conventional AV control groups. Data on 72 subjects (52 HeRO Graft and 20 AV graft controls) were obtained. The HeRO Graft and control cohorts were comparable in baseline characteristics. Adequacy of dialysis, bacteremia rates, and
adverse events were consistent between groups. Twelve month Kaplan–Meier estimates for primary and secondary patency rates were 34.8% and 67.6% in the HeRO Graft cohort, and 30.6% and 58.4% in the control cohort. There was no statistical difference in terms of patency between groups. The rates of intervention were 2.2/year for HeRO Graft and 1.6/year for the control (p = 0.100). Median days to loss of secondary patency was 238 for HeRO Graft versus 102 for the control (p = 0.032). The HeRO Graft appears to provide similar patency, adequacy of dialysis, and bacteremia rates to those of conventional AV grafts.
To date, there has only been one randomized trial involving the HeROâ Graft. The original clinical trial approved by the FDA for the IDE submission was a randomized study compared with conventional grafts. Feedback from investigators conducting the study indicated that the HeRO Graft might be better suited for patients who have exhausted peripheral venous access suitable for fistulas or grafts, so enrollment was stopped in favor of a study for catheter-dependent patients (1). Despite the small sample sizes, we feel that these data warrant presentation as they describe the only randomized study comparing the AV graft to the HeRO Graft. Venous outflow tract obstructive lesions, which have a documented history of rapid recurrence, are
the most frequent cause of conventional arteriovenous (AV) hemodialysis graft derangements (2). When venous obstruction cannot be resolved, the AV access is abandoned in favor of a new AV access. An alternative AV access in the ipsilateral upper extremity may not be possible with upper arm venous obstructive lesions. As a result, the contralateral upper extremity, or less favorably, the lower extremities are used. Due to the magnitude and implications of this problem, innovative solutions are highly desired to (a) minimize development of venous outflow obstruction and (b) bypass or correct obstructed venous outflow tracts. The hemodialysis reliable outflow (HeRO) Graft (Hemosphere, Inc., a CryoLife company) is a long-term subcutaneous AV graft with an outflow component that has been cleared by the Food and Drug Administration (FDA) for patients requiring long-term dialysis with obstructed venous outflow tracts unsuitable for conventional fistulas or grafts (3). The device consists of two components: a standard 6 mm inner diameter (ID)-expanded polytetrafluoroethylene (ePTFE) graft and a 5 mm ID nitinol-reinforced silicone single lumen outflow component. During implant in the upper extremities,
Address correspondence to: George M. Nassar, M.D.,1415 La Concha Lane, Houston, TX 77054, Tel.: 713-790-1963, Fax: 713-790-0766, or e-mail: [email protected]. Seminars in Dialysis—Vol 27, No 3 (May–June) 2014 pp. 310–318 DOI: 10.1111/sdi.12173 © 2014 Wiley Periodicals, Inc. 310
HERO GRAFT VERSUS CONVENTIONAL GRAFT
the graft is anastomosed to the brachial artery, tunneled subcutaneously, and attached via a titanium connector to the outflow component. The outflow component bypasses venous obstruction because it is tunneled from the connector to the internal jugular vein, where it enters the central venous circulation; the distal tip is positioned in the mid to upper right atrium. Similar to most AV grafts, the HeRO Graft requires tissue incorporation before it can be accessed for hemodialysis. During this bridging period, many patients will require a tunneled hemodialysis catheter. The HeRO Graft is cannulated in the same manner as a conventional AV ePTFE graft. In a separate, nonrandomized study in ESRD patients who were catheter-dependent due to venous outflow obstruction, performance of the HeRO Graft was better than the published performance of tunneled hemodialysis catheters in terms of bacteremia rates, patency and interventions (1). The study reported here evaluated HeRO Graft safety, performance, and long-term outcomes, relative to conventional AV ePTFE grafts in patients with vascular anatomies that are unable to support creation of AV fistula, but are able to support a conventional AV graft (i.e., graft-eligible patients). This study was registered with the National Institute of Health NCT00890045 and was conducted under an IDE # G030137 in support of the 510(k) submission to the Food and Drug Administration seeking market clearance for the HeRO Graft.
311
eligible patients, as determined by venous ultrasound mapping. The study commenced in July 2004, and the last follow-up visit occurred in April 2008. The original plan was that the first two subjects at each center to receive a HeRO Graft were considered training/roll-in subjects and their efficacy data were to be analyzed separately, to assess whether there is an implant learning curve possibly affecting outcomes. After two training cases, subsequent enrolled patients were randomized into a treatment group at each clinical site according to a blocked randomization schedule. Due to slow enrollment and investigator feedback, the study was terminated early, which reduced its statistical power. As a result, all HeRO Graft placements were grouped together and analyzed as a single cohort. Further discussion on this approach is provided in the statistical methods and discussion. Objectives Efficacy endpoints were primary and secondary patency, and the safety endpoint was serious device/ implant procedure-related adverse events. Additional ancillary data collected included all adverse events (including bloodstream infections, BSIs), device interventions, and adequacy of dialysis. Definitions Graft-Eligible Patient Definition
Methods Participants ESRD patients requiring dialysis who met the following criteria were eligible: age >21 years, brachial arteries >3 mm by duplex examination, life expectancy >2 years, and able to follow a daily aspirin and/or other oral anticoagulation/antiplatelet regimen. Patients were excluded for the following reasons: candidates for autologous AV fistula, bleeding diathesis or hypercoagulability, peripheral white blood cell count 0.999
94.2 28.9 8.1 36.5 63.5
95.0 28.0 4.9 25.0 75.0
>0.999 0.664 0.414 0.414
a Includes two enrolled subjects who did not receive the HeRO device. b Body mass index. In the control group, 17/20 subjects had an available BMI.
314
Nassar et al.
Patency and Intervention Data As shown in Table 3, there were no statistical differences between the HeRO Graft and control cohorts in terms of primary patency, assisted primary patency or secondary patency (log-rank p-values = 0.687, 0.737, and 0.656, respectively). Figures 1 and 2 illustrate Kaplan–Meier primary and secondary patency results for the HeRO Graft and the control cohort. For comparison, historic graft literature reports primary patency of 58% at 6 months and 42% at 12 months and secondary patency of 76% at 6 months and 65% at 12 months (8). The median time to loss of secondary patency for HeRO Graft was 238.0 days versus 102.5 days for the control group. The rates of intervention, 2.2 per year for the HeRO Graft and 1.6 per year for the control, were not statistically different (relative risk [95% confidence interval]: 1.3 [0.9, 1.9]; p = 0.100). The intervention results observed in this study were comparable to the graft literature intervention rates of 1.4-2.4 per year (9–12). Adequacy of Dialysis The median of days to first cannulation were 36.0 for the HeRO Graft and 34.5 for the control.
As presented in Table 4, the HeRO Graft mean Kt/V was 1.6 and URR was 72.8. The control cohort mean Kt/V was 1.7 and mean URR was 72.4. Adequacy of dialysis data for both the HeRO Graft and the control were comparable to the graft literature (Kt/V 1.37–1.62; URR 70.6–73.2) and within the preferred target range of the National Kidney Foundation Kidney Disease Outcomes Quality Initiative guidelines (Kt/V target > 1.4; URR target > 70) (13–18). The mean access blood flow rate was comparable between cohorts as assessed by duplex ultrasound at the 12-month follow-up visit. Adverse Events Serious device/implant procedure-related adverse events are presented in Table 5. A total of six (11.5%) HeRO Graft-related bleeding events occurred. In two of the six events, bleeding was indirectly related to the HeRO Graft implant procedure; in the first patient, coagulopathy was caused by other conditions and bleeding was not unexpected, and in the second patient, a heparin administration error occurred. In three of the six events, bleeding was directly attributed to an earlier generation 22 French HeRO Graft outflow component,
TABLE 2. Use of hemostasis inhibitors HeRO n (%) n Expecteda
Aspirin
52 48 45 44 42
10 (19.2) 11 (22.9) 10 (22.2) 11 (25) 9 (21.4)
Baseline 3 months 6 months 9 months 12 months
Plavix or generic 4 3 3 4 4
(7.7) (6.3) (6.7) (9.1) (9.5)
Control n (%) Anticoagulant 0 2 2 5 3
(0) (4.2) (4.4) (11.4) (7.1)
n Expecteda 20 19 19 17 15
Aspirin 4 5 5 3 3
(20) (26.3) (26.3) (17.6) (20)
Plavix or generic 0 1 1 1 1
(0) (5.3) (5.3) (5.9) (6.7)
Anticoagulant 2 1 1 1 1
(10) (5.3) (5.3) (5.9) (6.7)
a
Number of subjects who are not lost to follow-up, withdrew or died prior to that visit window close date.
TABLE 3. Patency and intervention results HeRO% [95% CI]d (n/N)e
Control% [95% CI]d (n/N)e
p-Value
47.0% [32.7–60.1] (23/49) 34.8% [21.9–48.0] (17/49)
48.2% [25.1–68.0] (9/19) 30.6% [11.8–51.9] (5/18)
0.687
86.2% [71.7–93.6] (30/36) 79.8% [63.1–89.5] (24/32)
88.8% [62.1–97.1] (13/15) 75.2% [46.1–90.0] (9/13)
0.737
76.7% [61.8–86.4] (34/45) 67.6% [52.0–79.1] (29/44)
58.4% [33.8–76.7] (11/19) 58.4% [33.8–76.7] (10/18)
0.656
118.0 (43) 88.5 (10) 238.0 (25) 2.2/year
114.5 (14) 183 (5) 102.5 (8) 1.6/year
0.442 0.624 0.032 0.100
a
Primary patency 6 month rates 12 month rates Assisted primary patencya 6 month rates 12 month rates Secondary patencya 6 month rates 12 month rates Median days to loss ofb Primary patency (n) Assisted primary patency (n) Secondary patency (n) Intervention ratesc a
p-values from log-rank test comparing HeRO with control. p-values from Kruskal–Wallis test comparing HeRO with control. Median calculated out of total patients who lost patency. p-values from Poisson regression analysis comparing HeRO with control. d Percent maintaining patency and 95% confidence interval at each time point based on Kaplan–Meier method. e “n/N” is the number left over the number at risk at each time point. b c
315
HERO GRAFT VERSUS CONVENTIONAL GRAFT
Fig. 1. Kaplan–Meier curve for freedom from loss of primary patency.
Two HeRO Graft adverse events in the same patient, classified as technical failures, involved devices believed to be damaged from vascular clamping during the implant procedure, requiring outflow component replacement. The outflow component was subsequently redesigned to reduce the risk of clamp damage. Two HeRO Graft embolisms were reported; one pulmonary embolism occurred in a patient who had an occluded device that was left in place for approximately 4 months, and the other embolism was in a patient who had an unknown patent foramen ovale and suffered a stroke after a declot procedure. The following six “other serious device/implant procedure-related adverse events” were reported for eight patients in the HeRO Graft cohort: hypotension with fever (the patient had previously undergone a gastrectomy and experienced ‘dumping syndrome’); ventricular tachycardia; cardiogenic shock; hyperkalemia due to a clotted access (n = 2); hypoxia postimplant procedure (n = 2); and, an elevated white blood cell count that did not result in infection (adjudicated as serious because the event delayed hospital discharge) . One control subject experienced two heart failure events as a result of repeated graft thrombosis and associated volume overload due to disruption in dialysis. The following “other device/implant procedurerelated adverse events” were reported for the control cohort: altered neurological function postimplant related to drug management, sleep deprivation and lack of dialysis; and postoperative chest pain. Bloodstream Infection
Fig. 2. Kaplan–Meier curve for freedom from loss of secondary patency.
which required an internal jugular venous cut-down. The sixth bleeding event was related to a HeRO Graft explant procedure. One HeRO Graft-related death occurred approximately 13 months postimplant as a result of sepsis (see additional detail under ‘Bloodstream Infection’).
Bloodstream infections (BSIs) are reported in Table 6. The overall BSI rate, regardless of device/ implant procedure relationship, in the HeRO Graft cohort was 0.72/1000 implant days (21.2%; 17 events in 11 subjects) and in the control cohort 0.62/ 1000 implant days (20.0%; three events in three subjects). The device/implant procedure-related BSI rate in the HeRO Graft cohort was 0.13/1000 implant days (5.8%; three events in three subjects) and in the control cohort was 0.21/1000 implant days (5.0%; two events in one subject). There was no statistical difference in the related BSI rates between groups (relative risk [95% confidence interval]: 0.6 [0.1, 3.8]; p = 0.623). The graft literature reports a rate of 0.11/1000 days (19), which is comparable to
TABLE 4. Adequacy of dialysis
Mean Kt/Vb (n) (SD; range) Mean URRc% (n) (SD; range) Mean access blood flow rate (n) (ml/minute) a
HeRO
Control
p-Valuea
1.6 (33) (0.3; 0.9–2.3) 72.8 (21) (6.0; 61.0–83.8) 1377.6 (38) (400.0; 716.4–2399.3)
1.7 (10) (0.4; 0.9–2.6) 72.4 (8) (6.3; 60.0–84.0) 1166.6 (10) (495.8; 563.8–2255.1)
0.5488 0.9945 0.1645
p-values from Student’s t-test comparing HeRO with control. K = dialyzer clearance of urea, t = dialysis time, and V = subject total body water. c Urea reduction ratio. b
316
Nassar et al.
the rates observed in the HeRO Graft and control cohorts. Discussion We believe this prospective randomized multicenter trial of AV graft-eligible dialysis patients is important for two main reasons. First, it is the only trial that attempted to prospectively evaluate the efficacy and safety of HeRO Graft compared with conventional AV grafts. Second, there are very few publications of prospective, randomized, multicenter clinical trials involving hemodialysis graft patient populations (8). Whether or not the HeRO Graft could minimize vascular access dysfunction and access loss compared with a conventional AV graft was not known at the time of conduct of the study. It was the intention of the trial to answer this question; however, due to slow enrollment and early termination of the study, a shortcoming of this trial, a final conclusion on equivalency could not be firmly established. Given these limitations, we still believe, the available data are important and valuable in a) primarily providing HeRO Graft outcomes in this prospective study with careful follow-up, and b) drawing comparisons between HeRO Graft and conventional AV graft outcomes using both the control group and available AV graft literature.
In reporting HeRO Graft outcomes, combining the HeRO Training/Roll-In and Treatment Subjects was only done after ensuring the comparability of these HeRO Graft subgroups. The advantage was to provide a larger cohort of HeRO Graft patients in whom measured outcomes can be reported. This eliminated the intention-to-treat design; however, as the randomized data were not sufficiently powered to answer the primary objectives of the study, we did not believe that this approach would create any tangible shortcomings. Even though no equivalency conclusions pertaining to the objectives of the study could be drawn from the data, 6- and 12-month primary and secondary patency rates in the HeRO Graft cohort were comparable to the conventional AV graft control group. There was a tendency to higher intervention rates with the HeRO cohort compared with the AV graft control group, but this did not reach statistical significance. However, the HeRO Graft cohort maintained secondary patency for nearly twice as long as the control group. This difference was statistically significant. We suspect, this result may be due to the elimination of a venous anastomosis with the HeRO Graft, which avoids complications associated with venous hyperplasia that the conventional AV grafts suffer from at their venous anastomosis. This unique property could be an advantage over conventional AV grafts that have the obligatory venous anastomosis. The latter commonly develops obstructive
TABLE 5. Serious device/implant procedure-related adverse events HeRO (%)a # eventsb/# subjectsc
Adverse event Bleeding, hemorrhage or hematoma Death Device technical failure Embolism Heart failure Infection (non-BSIe) Other Site pain Trauma to major veins, arteries, nerves Steal syndrome
11.5 1.9 1.9 3.8 0.0 3.8 9.6 1.9 1.9
Control (%)a # eventsb/# subjectsc
(6/6) (1/1) (2/1) (2/2) (0/0) (2/2) (8/5) (1/1) (1/1)
0.0 0.0 0.0 0.0 5.0 10.0 10.0 0.0 0.0
3.8 (2/2)
p-Valued
(0/0) (0/0) (0/0) (0/0) (2/1) (2/2) (2/2) (0/0) (0/0)
0.177 1.000 1.000 1.000 0.278 0.307 1.000 1.000 1.000
5.0 (1/1)
1.000
a
Percent of subjects with at least one event. b Total number of events. c Subjects with at least one event. d p-values from Fisher’s exact test. e Bloodstream infection. TABLE 6. Bloodstream infection (BSI) HeRO
All BSIsb Device/implant procedure-related BSIs Nonrelated BSIs a
Control
% (# events/# subjects)
Rate per 1000 implant days
% (# events/# subjects)
Rate per 1000 implant days
Relative risk [95% CIa]
21.2 (17/11) 5.8 (3/3)
0.72 0.13
20.0 (6/4) 5.0 (2/1)
0.62 0.21
1.2 [0.5, 2.9] 0.6 [0.1, 3.8]
19.2 (14/10)
0.62
15.0 (5/3)
0.52
1.2 [0.4, 3.3]
Confidence interval. b Contains data for all subjects, including those not implanted with the study device (two HeRO subjects).
317
HERO GRAFT VERSUS CONVENTIONAL GRAFT
venous stenosis that frequently propagates into the adjacent venous outflow tract. In addition, the HeRO Graft outflow component bypasses proximal venous occlusions subcutaneously and traverses stenosis within the central venous circulation. Although the HeRO Graft does not fail due to venous anastomosis stenosis, it can still fail due to other factors contributing to conventional AV graft failure, such as inadequate arterial inflow, hypotensive episodes, patient hypercoagulability, and steal syndrome. When compared with the conventional AV graft literature, at the time of conduct of the trial, the HeRO Graft cohort primary and secondary outcomes as well as intervention rates were comparable. Since the time these data were generated in the process of the approval of HeRO Graft by the FDA, and kept on file, additional data on HeRO Grafts have been reported in the literature. Of particular interest are three reports, all of which are retrospective, that provide outcomes data on additional 250 HeRO Graft implants (20–22). The primary and secondary HeRO Graft outcomes that emerged from these reports are comparable to our findings. In one of these reports, the HeRO Graft was compared with lower extremity conventional AV grafts in catheter-dependent population, and no difference in primary or secondary patency rates emerged (20). However, fewer interventions were required to maintain patency in the AV graft group (1.17 versus 2.21; p = 0.03) in this nonrandomized retrospective analysis in whom the HeRO Graft group had higher body mass index rates (20). In spite of that, the authors argue that they favor HeRO Grafts over lower extremity AV grafts in their catheter-dependent patients to preserve future lower extremity access options. They also argue that HeRO Graft is their preferred dialysis access in patients with obesity and peripheral vascular disease. We fully agree with their conclusions. The HeRO Graft-related BSI rates, reported in Table 6, were comparable to both the control AV graft group and the AV graft literature. Because HeRO Graft provides a long-term totally subcutaneous AV access that bypasses occlusion and traverses stenosis, it could be viewed as a solution to high BSI rates in patients who are catheter-dependent due to venous outflow stenosis. Therefore, the BSI rate associated with the HeRO Graft is of special interest. In this regard, the HeRO Graft BSI rates observed in the HeRO Graft cohort in this study are comparable to those noted with HeRO Grafts in a previous evaluation (1), and is much lower than the BSI rates of 0.7–6.5/1000 days reported in the literature with tunneled dialysis catheters shortly prior to the time of conduct of the study (6). Subsequently, Steerman (20) found HeRO Graft BSI rates to be comparable to conventional lower extremity AV grafts, and Gage (21) reported low HeRO Graft infection-related rates comparable to our findings. We attribute this apparently lower
BSI rate to the totally subcutaneous implant configuration of the HeRO Graft versus a catheter which is not totally subcutaneous. In conclusion, the patency, intervention, dialysis clearance, and BSI rates, observed in this prospective study, describe the outcomes of HeRO Graft implants in graft-eligible dialysis patients, and show that HeRO Graft outcomes are comparable to conventional AV grafts. These findings further support the conclusions of a previous report in which HeRO Graft was prospectively studied in a catheter-dependent patient population (1). The findings are also very comparable to those reported by other subsequent investigators in retrospective analyses of HeRO Graft outcomes (20–22). Our results should encourage further study and evaluation of the HeRO Graft in certain hemodialysis patient subpopulations who are struggling with their conventional vascular access due to safety, efficacy, or patency challenges. Acknowledgments The authors thank all the contributors (Barb Danson: Technical Writer; Andrea E. Fenton Abbs; Clinical Consultant; Eric D. Irwin: Clinical Event Committee Chair; William R. Omlie: Medical Advisor) for this study.
Disclosures Hemosphereâ, Inc., a CryoLife, Inc. company, the manufacturer of HeRO Graft, provided funding to conduct this FDA study. All of the authors’ institutions were compensated for conducting the study by Hemosphere. None of the physician authors received compensation for development of this manuscript. All trademarks used herein are owned by CryoLife or its subsidiaries.
References 1. Katzman HE, McLafferty RB, Ross JR, Glickman MH, Peden EK, Lawson JH: Initial experience and outcome of a new hemodialysis access device for catheter-dependent patients. J Vasc Surg 50:600–607, 2009. 2. Roy-Chaudhury P, Sukhatme VP, Cheung AK: Hemodialysis vascular access dysfunction: a cellular and molecular viewpoint. J Am Soc Nephrol 17:1112–1127, 2006. 3. Hemosphere, Inc. Instructions for use. Available at http://www.herograft.com/document-library. Accessed May 20, 2013 4. Sidawy AN, Gray R, Besarab A, Henry M, Ascher E, Silva M, Miller A, Scher L, Trerotola S, Gregory R, Rutherford R, Kent K: Recommended standards for reports dealing with arteriovenous hemodialysis access. J Vasc Surg 35:603–610, 2002. 5. Centers for Disease Control and Prevention Guidelines for the Prevention of Intravascular Catheter-Related Infections. MMWR, 51 (No.RR-10). 1–29, 2002 6. Nassar GM, Ayus JC: Infectious complications of the hemodialysis access. Kidney Int 60:1–13, 2001. 7. Nassar GM: Long-term performance of the hemodialysis reliable outflow device: the 56 month follow-up of the first clinical trial patient. Semin Dial 3:229–232, 2010. 8. Huber TS, Carter JW, Carter RL, Seeger J: Patency of autogenous and polytetrafluoroethylene upper extremity arteriovenous hemodialysis accesses: a systematic review. J Vasc Surg 38:1005–1011, 2003. 9. Bosman PJ, Blankestijn PJ, van der Graaf Y, Heintjes RJ, Koomans HA, Eikelboom BC: A comparison between PTFE and denatured homologous vein grafts for haemodialysis access: a prospective randomised multicentre trial. Eur J Vasc Endovasc Surg 16:126–132, 1998.
318
Nassar et al.
10. Madden RL, Lipkowitz GS, Browne BJ, Kurbanov A: A comparison of cryopreserved vein allografts and prosthetic grafts for hemodialysis access. Ann Vasc Surg 19:686–691, 2005. 11. Katzman HE, Glickman MH, Schild AF, Fujitani RM, Lawson JH: Multicenter evaluation of the bovine mesenteric vein bioprostheses for hemodialysis access in patients with an earlier failed prosthetic graft. J Am Coll Surg 201:223–230, 2005. 12. Hurlbert SN, Mattos MA, Henretta JP, Ramsey DE, Barkmeier LD, Hodgson KJ, Summer DS: Long-term patency rates, complications and cost-effectiveness of polytetrafluoroethylene (PTFE) grafts for hemodialysis access: a prospective study that compares Impra versus Gore-tex grafts. Cardiovasc Surg 6:652–656, 1998. 13. Sehgal AR, Snow RJ, Singer ME, Amini SB, DeOreo PB, Silver MR, Cebul RD: Barriers to adequate delivery of hemodialysis. Am J Kidney Dis 31:593–601, 1998. 14. Tonelli M, Muirhead N: Access type as a predictor of dialysis adequacy in chronic hemodialysis patients. ASAIO J 46:279–282, 2000. 15. Centers for Medicare and Medicaid Services, Kinney R. 2005 annual report: end stage renal disease clinical performance measures project. Am J Kidney Dis 48 (4 Suppl 2): S1–S106, 2006 16. Hemodialysis Adequacy 2006 Work Group. National Kidney Foundation KDOQI clinical practice guidelines for hemodialysis adequacy, update 2006. Am J Kidney Dis 48 (Suppl 1): S2–S90, 2006
17. Pastan S, Soucie JM, McClellan WM: Vascular access and increased risk of death among hemodialysis patients. Kidney Int 62:620–626, 2002. 18. Centers for Medicare & Medicaid Services. 2007 Annual Report, End Stage Renal Disease Clinical Performance Measures project. Department of Health and Human Services, Centers for Medicare & Medicaid Services, Office of Clinical Standards & Quality, Baltimore, Maryland, 2007 19. Hajjar J, Girard R, Marc JM, Ducruet L, Beruard M, Fadel B, Foret M, Lerda D, Roche C, Vallet M, Ayzac L, Fabry J: [Surveillance of infections in chronic hemodialysis patients (Article in French)]. Nephrologie 25:133–140, 2004. 20. Steerman SN, Wagner J, Higgins J, Kim C, Mirza A, Pavela J, Panneton JM, Glickman MH: Outcomes comparison of HeRO and lower extremity arteriovenous grafts in patients with long-standing renal failure. J Vasc Surg 57:776–783, 2013. 21. Gage SM, Katzman HE, Ross JR, Hohmann SE, Sharpe CA, Butterly DW, Lawson JH: Multi-center experience of 164 consecutive hemodialysis reliable outflow [HeRO] graft implants for hemodialysis treatment. Eur J Vasc Endovasc Surg 44:93–99, 2012. 22. Gebhard TA, Bryant JA, Grezaffi JA, Pabon-Ramos WM, Gage SM, Miller MJ, Husum KW, Suhocki PV, Sopko DR, Lawson JH, Smith TP, Kim CY: Percutaneous interventions on the hemodialysis reliable outflow vascular access device. J Vasc Interv Radiol 24:543– 549, 2013.
ABSTRACT Venous stenosis and occlusion are a major cause of vascular access dysfunction and failure. The HeRO Graft bypasses occlusion and traverses stenosis with outflow directly into the central venous circulation. A randomized, multicenter study was conducted to evaluate the efficacy and safety of the HeRO Graft relative to conventional AV grafts. The design was to enroll 143 patients in a 2:1 randomization ratio between HeRO and conventional AV control groups. Data on 72 subjects (52 HeRO Graft and 20 AV graft controls) were obtained. The HeRO Graft and control cohorts were comparable in baseline characteristics. Adequacy of dialysis, bacteremia rates, and
adverse events were consistent between groups. Twelve month Kaplan–Meier estimates for primary and secondary patency rates were 34.8% and 67.6% in the HeRO Graft cohort, and 30.6% and 58.4% in the control cohort. There was no statistical difference in terms of patency between groups. The rates of intervention were 2.2/year for HeRO Graft and 1.6/year for the control (p = 0.100). Median days to loss of secondary patency was 238 for HeRO Graft versus 102 for the control (p = 0.032). The HeRO Graft appears to provide similar patency, adequacy of dialysis, and bacteremia rates to those of conventional AV grafts.
To date, there has only been one randomized trial involving the HeROâ Graft. The original clinical trial approved by the FDA for the IDE submission was a randomized study compared with conventional grafts. Feedback from investigators conducting the study indicated that the HeRO Graft might be better suited for patients who have exhausted peripheral venous access suitable for fistulas or grafts, so enrollment was stopped in favor of a study for catheter-dependent patients (1). Despite the small sample sizes, we feel that these data warrant presentation as they describe the only randomized study comparing the AV graft to the HeRO Graft. Venous outflow tract obstructive lesions, which have a documented history of rapid recurrence, are
the most frequent cause of conventional arteriovenous (AV) hemodialysis graft derangements (2). When venous obstruction cannot be resolved, the AV access is abandoned in favor of a new AV access. An alternative AV access in the ipsilateral upper extremity may not be possible with upper arm venous obstructive lesions. As a result, the contralateral upper extremity, or less favorably, the lower extremities are used. Due to the magnitude and implications of this problem, innovative solutions are highly desired to (a) minimize development of venous outflow obstruction and (b) bypass or correct obstructed venous outflow tracts. The hemodialysis reliable outflow (HeRO) Graft (Hemosphere, Inc., a CryoLife company) is a long-term subcutaneous AV graft with an outflow component that has been cleared by the Food and Drug Administration (FDA) for patients requiring long-term dialysis with obstructed venous outflow tracts unsuitable for conventional fistulas or grafts (3). The device consists of two components: a standard 6 mm inner diameter (ID)-expanded polytetrafluoroethylene (ePTFE) graft and a 5 mm ID nitinol-reinforced silicone single lumen outflow component. During implant in the upper extremities,
Address correspondence to: George M. Nassar, M.D.,1415 La Concha Lane, Houston, TX 77054, Tel.: 713-790-1963, Fax: 713-790-0766, or e-mail: [email protected]. Seminars in Dialysis—Vol 27, No 3 (May–June) 2014 pp. 310–318 DOI: 10.1111/sdi.12173 © 2014 Wiley Periodicals, Inc. 310
HERO GRAFT VERSUS CONVENTIONAL GRAFT
the graft is anastomosed to the brachial artery, tunneled subcutaneously, and attached via a titanium connector to the outflow component. The outflow component bypasses venous obstruction because it is tunneled from the connector to the internal jugular vein, where it enters the central venous circulation; the distal tip is positioned in the mid to upper right atrium. Similar to most AV grafts, the HeRO Graft requires tissue incorporation before it can be accessed for hemodialysis. During this bridging period, many patients will require a tunneled hemodialysis catheter. The HeRO Graft is cannulated in the same manner as a conventional AV ePTFE graft. In a separate, nonrandomized study in ESRD patients who were catheter-dependent due to venous outflow obstruction, performance of the HeRO Graft was better than the published performance of tunneled hemodialysis catheters in terms of bacteremia rates, patency and interventions (1). The study reported here evaluated HeRO Graft safety, performance, and long-term outcomes, relative to conventional AV ePTFE grafts in patients with vascular anatomies that are unable to support creation of AV fistula, but are able to support a conventional AV graft (i.e., graft-eligible patients). This study was registered with the National Institute of Health NCT00890045 and was conducted under an IDE # G030137 in support of the 510(k) submission to the Food and Drug Administration seeking market clearance for the HeRO Graft.
311
eligible patients, as determined by venous ultrasound mapping. The study commenced in July 2004, and the last follow-up visit occurred in April 2008. The original plan was that the first two subjects at each center to receive a HeRO Graft were considered training/roll-in subjects and their efficacy data were to be analyzed separately, to assess whether there is an implant learning curve possibly affecting outcomes. After two training cases, subsequent enrolled patients were randomized into a treatment group at each clinical site according to a blocked randomization schedule. Due to slow enrollment and investigator feedback, the study was terminated early, which reduced its statistical power. As a result, all HeRO Graft placements were grouped together and analyzed as a single cohort. Further discussion on this approach is provided in the statistical methods and discussion. Objectives Efficacy endpoints were primary and secondary patency, and the safety endpoint was serious device/ implant procedure-related adverse events. Additional ancillary data collected included all adverse events (including bloodstream infections, BSIs), device interventions, and adequacy of dialysis. Definitions Graft-Eligible Patient Definition
Methods Participants ESRD patients requiring dialysis who met the following criteria were eligible: age >21 years, brachial arteries >3 mm by duplex examination, life expectancy >2 years, and able to follow a daily aspirin and/or other oral anticoagulation/antiplatelet regimen. Patients were excluded for the following reasons: candidates for autologous AV fistula, bleeding diathesis or hypercoagulability, peripheral white blood cell count 0.999
94.2 28.9 8.1 36.5 63.5
95.0 28.0 4.9 25.0 75.0
>0.999 0.664 0.414 0.414
a Includes two enrolled subjects who did not receive the HeRO device. b Body mass index. In the control group, 17/20 subjects had an available BMI.
314
Nassar et al.
Patency and Intervention Data As shown in Table 3, there were no statistical differences between the HeRO Graft and control cohorts in terms of primary patency, assisted primary patency or secondary patency (log-rank p-values = 0.687, 0.737, and 0.656, respectively). Figures 1 and 2 illustrate Kaplan–Meier primary and secondary patency results for the HeRO Graft and the control cohort. For comparison, historic graft literature reports primary patency of 58% at 6 months and 42% at 12 months and secondary patency of 76% at 6 months and 65% at 12 months (8). The median time to loss of secondary patency for HeRO Graft was 238.0 days versus 102.5 days for the control group. The rates of intervention, 2.2 per year for the HeRO Graft and 1.6 per year for the control, were not statistically different (relative risk [95% confidence interval]: 1.3 [0.9, 1.9]; p = 0.100). The intervention results observed in this study were comparable to the graft literature intervention rates of 1.4-2.4 per year (9–12). Adequacy of Dialysis The median of days to first cannulation were 36.0 for the HeRO Graft and 34.5 for the control.
As presented in Table 4, the HeRO Graft mean Kt/V was 1.6 and URR was 72.8. The control cohort mean Kt/V was 1.7 and mean URR was 72.4. Adequacy of dialysis data for both the HeRO Graft and the control were comparable to the graft literature (Kt/V 1.37–1.62; URR 70.6–73.2) and within the preferred target range of the National Kidney Foundation Kidney Disease Outcomes Quality Initiative guidelines (Kt/V target > 1.4; URR target > 70) (13–18). The mean access blood flow rate was comparable between cohorts as assessed by duplex ultrasound at the 12-month follow-up visit. Adverse Events Serious device/implant procedure-related adverse events are presented in Table 5. A total of six (11.5%) HeRO Graft-related bleeding events occurred. In two of the six events, bleeding was indirectly related to the HeRO Graft implant procedure; in the first patient, coagulopathy was caused by other conditions and bleeding was not unexpected, and in the second patient, a heparin administration error occurred. In three of the six events, bleeding was directly attributed to an earlier generation 22 French HeRO Graft outflow component,
TABLE 2. Use of hemostasis inhibitors HeRO n (%) n Expecteda
Aspirin
52 48 45 44 42
10 (19.2) 11 (22.9) 10 (22.2) 11 (25) 9 (21.4)
Baseline 3 months 6 months 9 months 12 months
Plavix or generic 4 3 3 4 4
(7.7) (6.3) (6.7) (9.1) (9.5)
Control n (%) Anticoagulant 0 2 2 5 3
(0) (4.2) (4.4) (11.4) (7.1)
n Expecteda 20 19 19 17 15
Aspirin 4 5 5 3 3
(20) (26.3) (26.3) (17.6) (20)
Plavix or generic 0 1 1 1 1
(0) (5.3) (5.3) (5.9) (6.7)
Anticoagulant 2 1 1 1 1
(10) (5.3) (5.3) (5.9) (6.7)
a
Number of subjects who are not lost to follow-up, withdrew or died prior to that visit window close date.
TABLE 3. Patency and intervention results HeRO% [95% CI]d (n/N)e
Control% [95% CI]d (n/N)e
p-Value
47.0% [32.7–60.1] (23/49) 34.8% [21.9–48.0] (17/49)
48.2% [25.1–68.0] (9/19) 30.6% [11.8–51.9] (5/18)
0.687
86.2% [71.7–93.6] (30/36) 79.8% [63.1–89.5] (24/32)
88.8% [62.1–97.1] (13/15) 75.2% [46.1–90.0] (9/13)
0.737
76.7% [61.8–86.4] (34/45) 67.6% [52.0–79.1] (29/44)
58.4% [33.8–76.7] (11/19) 58.4% [33.8–76.7] (10/18)
0.656
118.0 (43) 88.5 (10) 238.0 (25) 2.2/year
114.5 (14) 183 (5) 102.5 (8) 1.6/year
0.442 0.624 0.032 0.100
a
Primary patency 6 month rates 12 month rates Assisted primary patencya 6 month rates 12 month rates Secondary patencya 6 month rates 12 month rates Median days to loss ofb Primary patency (n) Assisted primary patency (n) Secondary patency (n) Intervention ratesc a
p-values from log-rank test comparing HeRO with control. p-values from Kruskal–Wallis test comparing HeRO with control. Median calculated out of total patients who lost patency. p-values from Poisson regression analysis comparing HeRO with control. d Percent maintaining patency and 95% confidence interval at each time point based on Kaplan–Meier method. e “n/N” is the number left over the number at risk at each time point. b c
315
HERO GRAFT VERSUS CONVENTIONAL GRAFT
Fig. 1. Kaplan–Meier curve for freedom from loss of primary patency.
Two HeRO Graft adverse events in the same patient, classified as technical failures, involved devices believed to be damaged from vascular clamping during the implant procedure, requiring outflow component replacement. The outflow component was subsequently redesigned to reduce the risk of clamp damage. Two HeRO Graft embolisms were reported; one pulmonary embolism occurred in a patient who had an occluded device that was left in place for approximately 4 months, and the other embolism was in a patient who had an unknown patent foramen ovale and suffered a stroke after a declot procedure. The following six “other serious device/implant procedure-related adverse events” were reported for eight patients in the HeRO Graft cohort: hypotension with fever (the patient had previously undergone a gastrectomy and experienced ‘dumping syndrome’); ventricular tachycardia; cardiogenic shock; hyperkalemia due to a clotted access (n = 2); hypoxia postimplant procedure (n = 2); and, an elevated white blood cell count that did not result in infection (adjudicated as serious because the event delayed hospital discharge) . One control subject experienced two heart failure events as a result of repeated graft thrombosis and associated volume overload due to disruption in dialysis. The following “other device/implant procedurerelated adverse events” were reported for the control cohort: altered neurological function postimplant related to drug management, sleep deprivation and lack of dialysis; and postoperative chest pain. Bloodstream Infection
Fig. 2. Kaplan–Meier curve for freedom from loss of secondary patency.
which required an internal jugular venous cut-down. The sixth bleeding event was related to a HeRO Graft explant procedure. One HeRO Graft-related death occurred approximately 13 months postimplant as a result of sepsis (see additional detail under ‘Bloodstream Infection’).
Bloodstream infections (BSIs) are reported in Table 6. The overall BSI rate, regardless of device/ implant procedure relationship, in the HeRO Graft cohort was 0.72/1000 implant days (21.2%; 17 events in 11 subjects) and in the control cohort 0.62/ 1000 implant days (20.0%; three events in three subjects). The device/implant procedure-related BSI rate in the HeRO Graft cohort was 0.13/1000 implant days (5.8%; three events in three subjects) and in the control cohort was 0.21/1000 implant days (5.0%; two events in one subject). There was no statistical difference in the related BSI rates between groups (relative risk [95% confidence interval]: 0.6 [0.1, 3.8]; p = 0.623). The graft literature reports a rate of 0.11/1000 days (19), which is comparable to
TABLE 4. Adequacy of dialysis
Mean Kt/Vb (n) (SD; range) Mean URRc% (n) (SD; range) Mean access blood flow rate (n) (ml/minute) a
HeRO
Control
p-Valuea
1.6 (33) (0.3; 0.9–2.3) 72.8 (21) (6.0; 61.0–83.8) 1377.6 (38) (400.0; 716.4–2399.3)
1.7 (10) (0.4; 0.9–2.6) 72.4 (8) (6.3; 60.0–84.0) 1166.6 (10) (495.8; 563.8–2255.1)
0.5488 0.9945 0.1645
p-values from Student’s t-test comparing HeRO with control. K = dialyzer clearance of urea, t = dialysis time, and V = subject total body water. c Urea reduction ratio. b
316
Nassar et al.
the rates observed in the HeRO Graft and control cohorts. Discussion We believe this prospective randomized multicenter trial of AV graft-eligible dialysis patients is important for two main reasons. First, it is the only trial that attempted to prospectively evaluate the efficacy and safety of HeRO Graft compared with conventional AV grafts. Second, there are very few publications of prospective, randomized, multicenter clinical trials involving hemodialysis graft patient populations (8). Whether or not the HeRO Graft could minimize vascular access dysfunction and access loss compared with a conventional AV graft was not known at the time of conduct of the study. It was the intention of the trial to answer this question; however, due to slow enrollment and early termination of the study, a shortcoming of this trial, a final conclusion on equivalency could not be firmly established. Given these limitations, we still believe, the available data are important and valuable in a) primarily providing HeRO Graft outcomes in this prospective study with careful follow-up, and b) drawing comparisons between HeRO Graft and conventional AV graft outcomes using both the control group and available AV graft literature.
In reporting HeRO Graft outcomes, combining the HeRO Training/Roll-In and Treatment Subjects was only done after ensuring the comparability of these HeRO Graft subgroups. The advantage was to provide a larger cohort of HeRO Graft patients in whom measured outcomes can be reported. This eliminated the intention-to-treat design; however, as the randomized data were not sufficiently powered to answer the primary objectives of the study, we did not believe that this approach would create any tangible shortcomings. Even though no equivalency conclusions pertaining to the objectives of the study could be drawn from the data, 6- and 12-month primary and secondary patency rates in the HeRO Graft cohort were comparable to the conventional AV graft control group. There was a tendency to higher intervention rates with the HeRO cohort compared with the AV graft control group, but this did not reach statistical significance. However, the HeRO Graft cohort maintained secondary patency for nearly twice as long as the control group. This difference was statistically significant. We suspect, this result may be due to the elimination of a venous anastomosis with the HeRO Graft, which avoids complications associated with venous hyperplasia that the conventional AV grafts suffer from at their venous anastomosis. This unique property could be an advantage over conventional AV grafts that have the obligatory venous anastomosis. The latter commonly develops obstructive
TABLE 5. Serious device/implant procedure-related adverse events HeRO (%)a # eventsb/# subjectsc
Adverse event Bleeding, hemorrhage or hematoma Death Device technical failure Embolism Heart failure Infection (non-BSIe) Other Site pain Trauma to major veins, arteries, nerves Steal syndrome
11.5 1.9 1.9 3.8 0.0 3.8 9.6 1.9 1.9
Control (%)a # eventsb/# subjectsc
(6/6) (1/1) (2/1) (2/2) (0/0) (2/2) (8/5) (1/1) (1/1)
0.0 0.0 0.0 0.0 5.0 10.0 10.0 0.0 0.0
3.8 (2/2)
p-Valued
(0/0) (0/0) (0/0) (0/0) (2/1) (2/2) (2/2) (0/0) (0/0)
0.177 1.000 1.000 1.000 0.278 0.307 1.000 1.000 1.000
5.0 (1/1)
1.000
a
Percent of subjects with at least one event. b Total number of events. c Subjects with at least one event. d p-values from Fisher’s exact test. e Bloodstream infection. TABLE 6. Bloodstream infection (BSI) HeRO
All BSIsb Device/implant procedure-related BSIs Nonrelated BSIs a
Control
% (# events/# subjects)
Rate per 1000 implant days
% (# events/# subjects)
Rate per 1000 implant days
Relative risk [95% CIa]
21.2 (17/11) 5.8 (3/3)
0.72 0.13
20.0 (6/4) 5.0 (2/1)
0.62 0.21
1.2 [0.5, 2.9] 0.6 [0.1, 3.8]
19.2 (14/10)
0.62
15.0 (5/3)
0.52
1.2 [0.4, 3.3]
Confidence interval. b Contains data for all subjects, including those not implanted with the study device (two HeRO subjects).
317
HERO GRAFT VERSUS CONVENTIONAL GRAFT
venous stenosis that frequently propagates into the adjacent venous outflow tract. In addition, the HeRO Graft outflow component bypasses proximal venous occlusions subcutaneously and traverses stenosis within the central venous circulation. Although the HeRO Graft does not fail due to venous anastomosis stenosis, it can still fail due to other factors contributing to conventional AV graft failure, such as inadequate arterial inflow, hypotensive episodes, patient hypercoagulability, and steal syndrome. When compared with the conventional AV graft literature, at the time of conduct of the trial, the HeRO Graft cohort primary and secondary outcomes as well as intervention rates were comparable. Since the time these data were generated in the process of the approval of HeRO Graft by the FDA, and kept on file, additional data on HeRO Grafts have been reported in the literature. Of particular interest are three reports, all of which are retrospective, that provide outcomes data on additional 250 HeRO Graft implants (20–22). The primary and secondary HeRO Graft outcomes that emerged from these reports are comparable to our findings. In one of these reports, the HeRO Graft was compared with lower extremity conventional AV grafts in catheter-dependent population, and no difference in primary or secondary patency rates emerged (20). However, fewer interventions were required to maintain patency in the AV graft group (1.17 versus 2.21; p = 0.03) in this nonrandomized retrospective analysis in whom the HeRO Graft group had higher body mass index rates (20). In spite of that, the authors argue that they favor HeRO Grafts over lower extremity AV grafts in their catheter-dependent patients to preserve future lower extremity access options. They also argue that HeRO Graft is their preferred dialysis access in patients with obesity and peripheral vascular disease. We fully agree with their conclusions. The HeRO Graft-related BSI rates, reported in Table 6, were comparable to both the control AV graft group and the AV graft literature. Because HeRO Graft provides a long-term totally subcutaneous AV access that bypasses occlusion and traverses stenosis, it could be viewed as a solution to high BSI rates in patients who are catheter-dependent due to venous outflow stenosis. Therefore, the BSI rate associated with the HeRO Graft is of special interest. In this regard, the HeRO Graft BSI rates observed in the HeRO Graft cohort in this study are comparable to those noted with HeRO Grafts in a previous evaluation (1), and is much lower than the BSI rates of 0.7–6.5/1000 days reported in the literature with tunneled dialysis catheters shortly prior to the time of conduct of the study (6). Subsequently, Steerman (20) found HeRO Graft BSI rates to be comparable to conventional lower extremity AV grafts, and Gage (21) reported low HeRO Graft infection-related rates comparable to our findings. We attribute this apparently lower
BSI rate to the totally subcutaneous implant configuration of the HeRO Graft versus a catheter which is not totally subcutaneous. In conclusion, the patency, intervention, dialysis clearance, and BSI rates, observed in this prospective study, describe the outcomes of HeRO Graft implants in graft-eligible dialysis patients, and show that HeRO Graft outcomes are comparable to conventional AV grafts. These findings further support the conclusions of a previous report in which HeRO Graft was prospectively studied in a catheter-dependent patient population (1). The findings are also very comparable to those reported by other subsequent investigators in retrospective analyses of HeRO Graft outcomes (20–22). Our results should encourage further study and evaluation of the HeRO Graft in certain hemodialysis patient subpopulations who are struggling with their conventional vascular access due to safety, efficacy, or patency challenges. Acknowledgments The authors thank all the contributors (Barb Danson: Technical Writer; Andrea E. Fenton Abbs; Clinical Consultant; Eric D. Irwin: Clinical Event Committee Chair; William R. Omlie: Medical Advisor) for this study.
Disclosures Hemosphereâ, Inc., a CryoLife, Inc. company, the manufacturer of HeRO Graft, provided funding to conduct this FDA study. All of the authors’ institutions were compensated for conducting the study by Hemosphere. None of the physician authors received compensation for development of this manuscript. All trademarks used herein are owned by CryoLife or its subsidiaries.
References 1. Katzman HE, McLafferty RB, Ross JR, Glickman MH, Peden EK, Lawson JH: Initial experience and outcome of a new hemodialysis access device for catheter-dependent patients. J Vasc Surg 50:600–607, 2009. 2. Roy-Chaudhury P, Sukhatme VP, Cheung AK: Hemodialysis vascular access dysfunction: a cellular and molecular viewpoint. J Am Soc Nephrol 17:1112–1127, 2006. 3. Hemosphere, Inc. Instructions for use. Available at http://www.herograft.com/document-library. Accessed May 20, 2013 4. Sidawy AN, Gray R, Besarab A, Henry M, Ascher E, Silva M, Miller A, Scher L, Trerotola S, Gregory R, Rutherford R, Kent K: Recommended standards for reports dealing with arteriovenous hemodialysis access. J Vasc Surg 35:603–610, 2002. 5. Centers for Disease Control and Prevention Guidelines for the Prevention of Intravascular Catheter-Related Infections. MMWR, 51 (No.RR-10). 1–29, 2002 6. Nassar GM, Ayus JC: Infectious complications of the hemodialysis access. Kidney Int 60:1–13, 2001. 7. Nassar GM: Long-term performance of the hemodialysis reliable outflow device: the 56 month follow-up of the first clinical trial patient. Semin Dial 3:229–232, 2010. 8. Huber TS, Carter JW, Carter RL, Seeger J: Patency of autogenous and polytetrafluoroethylene upper extremity arteriovenous hemodialysis accesses: a systematic review. J Vasc Surg 38:1005–1011, 2003. 9. Bosman PJ, Blankestijn PJ, van der Graaf Y, Heintjes RJ, Koomans HA, Eikelboom BC: A comparison between PTFE and denatured homologous vein grafts for haemodialysis access: a prospective randomised multicentre trial. Eur J Vasc Endovasc Surg 16:126–132, 1998.
318
Nassar et al.
10. Madden RL, Lipkowitz GS, Browne BJ, Kurbanov A: A comparison of cryopreserved vein allografts and prosthetic grafts for hemodialysis access. Ann Vasc Surg 19:686–691, 2005. 11. Katzman HE, Glickman MH, Schild AF, Fujitani RM, Lawson JH: Multicenter evaluation of the bovine mesenteric vein bioprostheses for hemodialysis access in patients with an earlier failed prosthetic graft. J Am Coll Surg 201:223–230, 2005. 12. Hurlbert SN, Mattos MA, Henretta JP, Ramsey DE, Barkmeier LD, Hodgson KJ, Summer DS: Long-term patency rates, complications and cost-effectiveness of polytetrafluoroethylene (PTFE) grafts for hemodialysis access: a prospective study that compares Impra versus Gore-tex grafts. Cardiovasc Surg 6:652–656, 1998. 13. Sehgal AR, Snow RJ, Singer ME, Amini SB, DeOreo PB, Silver MR, Cebul RD: Barriers to adequate delivery of hemodialysis. Am J Kidney Dis 31:593–601, 1998. 14. Tonelli M, Muirhead N: Access type as a predictor of dialysis adequacy in chronic hemodialysis patients. ASAIO J 46:279–282, 2000. 15. Centers for Medicare and Medicaid Services, Kinney R. 2005 annual report: end stage renal disease clinical performance measures project. Am J Kidney Dis 48 (4 Suppl 2): S1–S106, 2006 16. Hemodialysis Adequacy 2006 Work Group. National Kidney Foundation KDOQI clinical practice guidelines for hemodialysis adequacy, update 2006. Am J Kidney Dis 48 (Suppl 1): S2–S90, 2006
17. Pastan S, Soucie JM, McClellan WM: Vascular access and increased risk of death among hemodialysis patients. Kidney Int 62:620–626, 2002. 18. Centers for Medicare & Medicaid Services. 2007 Annual Report, End Stage Renal Disease Clinical Performance Measures project. Department of Health and Human Services, Centers for Medicare & Medicaid Services, Office of Clinical Standards & Quality, Baltimore, Maryland, 2007 19. Hajjar J, Girard R, Marc JM, Ducruet L, Beruard M, Fadel B, Foret M, Lerda D, Roche C, Vallet M, Ayzac L, Fabry J: [Surveillance of infections in chronic hemodialysis patients (Article in French)]. Nephrologie 25:133–140, 2004. 20. Steerman SN, Wagner J, Higgins J, Kim C, Mirza A, Pavela J, Panneton JM, Glickman MH: Outcomes comparison of HeRO and lower extremity arteriovenous grafts in patients with long-standing renal failure. J Vasc Surg 57:776–783, 2013. 21. Gage SM, Katzman HE, Ross JR, Hohmann SE, Sharpe CA, Butterly DW, Lawson JH: Multi-center experience of 164 consecutive hemodialysis reliable outflow [HeRO] graft implants for hemodialysis treatment. Eur J Vasc Endovasc Surg 44:93–99, 2012. 22. Gebhard TA, Bryant JA, Grezaffi JA, Pabon-Ramos WM, Gage SM, Miller MJ, Husum KW, Suhocki PV, Sopko DR, Lawson JH, Smith TP, Kim CY: Percutaneous interventions on the hemodialysis reliable outflow vascular access device. J Vasc Interv Radiol 24:543– 549, 2013.
