European Journal of Vascular & Endovascular Surgery
Volume 31, Issue 2 , Pages 181-186, February 2006

Depopulated Bovine Ureteric Xenograft for Complex Haemodialysis Vascular Access

  • C.R. Darby

      Affiliations

    • Oxford Transplant Centre, Headington, Oxford, UK
    • Department of Vascular Surgery, Headington, Oxford, UK
    • Corresponding Author InformationCorresponding author. Mr C.R. Darby, MBBS, FRCS, MD, Consultant Vascular and Transplant Surgeon, Oxford Transplant Centre, Churchill Hospital, Old Road, Headington, Oxford OX3 7LJ, UK.
    • ,
  • D. Roy

      Affiliations

    • Oxford Transplant Centre, Headington, Oxford, UK
    • ,
  • D. Deardon

      Affiliations

    • Oxford Transplant Centre, Headington, Oxford, UK
    • ,
  • A. Cornall

      Affiliations

    • Oxford Transplant Centre, Headington, Oxford, UK

Accepted 6 July 2005. published online 30 August 2005.

Article Outline

Abstract 

Objectives

To assess performance of a decellularised bovine ureter vascular graft for haemodialysis in patients for whom conventional access was not possible.

Methods

The SynerGraft® Vascular Graft Model 100 (SGVG 100) is a bovine ureter modified by a tissue-engineering depopulation technology and uniquely it is not chemically cross-linked. SGVG 100 was implanted in patients with a failed fistula or vascular access grafts. Graft patency was the primary outcome; secondary outcomes included adverse events and associated treatments.

Results

25 SGVG 100 were implanted in 23 patients; mean age was 59±14 years. Mean follow-up was 370 days. The mean time to occlusion (19 events) was 215±141 days with patency re-established in 14 of 18 surgical interventions. Thirty angioplasties were performed on 14 SGVG 100 for luminal/anastomotic stenosis. Two grafts demonstrated areas of dilation; however, both grafts continue to be usable at last reported follow-up (930 and 602 days) with no further changes in graft size. Primary patency, assisted primary patency, secondary patency, and freedom from infection were 29, 45, 81, and 95% at 1 year, respectively.

Conclusions

This report demonstrates SGVG 100 is a stable vascular access conduit, providing a suitable graft alternative when autologous vein is not available.

Keywords: Vascular access, Bovine ureter, Tissue engineering, Surgery, Outcome

 

Back to Article Outline

1. Introduction 

The incidence of end stage renal failure treated by dialysis continues to increase in both Europe and the USA.1 Haemodialysis is the most common modality for dialysis, comprising 92% of new patients that are treated in the USA.2 The gold standard for vascular access is the creation of a primary arteriovenous fistula (AVF) using native vein. In Europe and more recently in America the benefits of this approach have been fully recognised and this approach is formalised by the DOQI guidelines in the USA.3 However, it is recognised that there will always be a percentage of patients for whom use of native vein is not possible. Hence, there is an extensive history in the use of synthetic arterio venous grafts (AVG) for this purpose. The most widely used bridge graft material for haemodialysis is polytetrafluoroethylene (PTFE).4, 5 PTFE has been shown to have a wide range of primary patency rates of between 41 and 68% at 1 year.5

The main complications for PTFE grafts used for haemodialysis access have been thrombosis, reported to contribute to 70% of the failures6, 7 and infection, which on average occurs in 10% of implanted PTFE grafts.5, 7, 8, 9

In the past before the widespread use of synthetic grafts, biological grafts were used for vascular access and peripheral bypass surgery. Cryopreserved vascular allograft tissue has been shown to be useful in difficult vascular access situations.10, 11 Cryopreserved vascular tissue is effective in cases of active infection and can be placed directly in the infected field with a low recurrence of infection.10, 11, 12

Xenografts, including natural bovine tissues (carotid artery, Artegraft® and visceral veins, Procol® and ovine collagen tubes) have been used for haemodialysis access. These grafts are traditionally treated with glutaraldehyde cross-linking to stabilise the connective tissue matrix. Reported results from clinical studies have been mixed with failures related to aneurysm formation, calcification and infection.13, 14, 15, 16, 17, 18, 19

The graft described here is unlike any other previously implanted in that it is decellularised to remove antigenicity and there is no chemical treatment allowing neo cellularisation by the host and the retention of some of the native biomechanical characteristics. To generate this graft a new method to process xenograft tissue was developed, SynerGraft® Vascular Graft Model 100 (SGVG 100, CryoLife, Inc., Kennesaw, GA, USA). The cellular elements have been removed using a proprietary SynerGraft antigen reduction technology. This technology was developed as an alternative to chemical cross-linking to render biological grafts non-immunogenic. Preliminary studies demonstrated removal of cells with retention of natural biomechanical characteristics (compliance, longitudinal and circumferential tensile strength).20

There has been no previously published data in humans using this graft for vascular access. However, in a dog arteriovenous shunt model, the primary patency of the SGVG 100 at 6 months was 72.6% (versus 58.6% for a PTFE conduits) and the 12-month patency was no different from PTFE (58.6 and 57.4%, respectively). There were no aneurysms in the SGVG 100 grafts. There were no infections among the SGVG 100 grafts, despite twice weekly sham access, whereas three of the PTFE implants became infected with Staphylococcus intermedius (epidermidis). Migration of interstitial cells into the previously acellular matrix was ongoing as early as 2.5 months, with additional re-cellularisation and collagen biosynthetic activity observed at 1 year.21

These preclinical studies supported the application of the SGVG 100 conduit for clinical use in humans. Hence, this single centre non-randomised study was performed to review the performance of this novel biological graft.

Back to Article Outline

2. Materials and Methods 

The bovine ureter (SGVG 100) is obtained from cows in the USA. The graft is processed by hypotonic water lysis and incubation with DNAse and RNAse to destroy the cellular component, which are removed by isotonic washing over a 12-day period. The graft is gamma irradiated to sterilise it. The graft is available in 25, 35 and 50cm lengths of approximately 0.7cm internal diameter packed in pouches of sterile saline. The graft can be kept at room temperature with a 2-year shelf life.

2.1. Patients 

Patients were deemed suitable for treatment with a bovine graft if multiple attempts to fashion vascular access had been made previously and attempts at autogenous venous access was not likely to be successful (Table 1). Access in the lower limb was created only in patients with an occluded superior vena cava. No special consent was required as the product was licensed in the UK for this indication. However, a detailed patient information leaflet describing the product and any potential problems was provided and consent was taken from each patient for recording of all demographic, surgical and outcome data.

Table 1. Patient demographics
Analysis parameterSGVG 100 grafts (n=25)*
Age (mean ± standard deviation)59.0±14.3 years
Age range (minimum to maximum)35.0 to 81.0 years
Gender8 male, 17 female
Medical history
Diabetes5
Hypertension13
Peripheral vascular disease5
Coronary artery disease5
Smoking history (current or previous)11
Polycystic kidneys4
Chronic pyelonephritis3

Previous procedures on limb (mean±standard deviation)

2.4±1.0
02
12
26
3+15

Years on haemodialysis (mean±standard deviation)

4.2±1.6 years
0–1 year2
2–3 years9
4–5 years7
6+ years7

Indication for surgery†

Primary access1
Failed fistula21
Failed graft9
Infection1

*Twenty-five grafts implanted in 23 patients (patients with two grafts are treated as independent entries in the analysis).

†May have more than one entry per graft.

2.2. Techniques 

All patients received their access graft under a general anaesthetic. Patients received a single dose of antibiotic prophylaxis (cefuroxime). No heparin or steroids was given. The brachial or femoral artery was prepared and clamped. An end graft to side recipient native vessel anastomosis was prepared with no undue handling and no clamping of the graft allowed. The venous site was prepared and the graft tunnelled avoiding traction injury. The graft was pulled down a hollow tunneller to achieve this. The graft was anastomosed to the vein end to side. Clamps were removed from the vein and artery and flow restored.

2.3. Data and statistics 

The data was entered prospectively onto an online database. Outcome data was recorded, determining operative technique, primary patency, assisted primary patency and secondary patency, morbidity and mortality. Endpoints were set according to the reporting standards for arterio-venous access of the Society for Vascular Surgery and the American Association for Vascular Surgery.22 A data entry form was completed prospectively at the following time points, preoperatively, at the time of surgery, at each post operative follow up appointment, at the time of graft complication, at the time of color duplex assessment or flow monitoring and at patient death. Data was analysed by Kaplan–Meier statistical analysis for the calculation of overall freedom from events.

Back to Article Outline

3. Results 

The range of follow-up time was 44–930 days, with a mean follow-up time of 370±256 days and total follow-up time of 9247 days (25.3 patient-years). Follow-up data collection was 100% complete. Of the 25 SGVG 100 implanted, 22 went on to undergo dialysis access (one occluded day 44 and two are currently awaiting first dialysis—day 46 and day 48). The range of time to first access was 21–148 days (mean 67 days, median 55 days). Intended first needling was at 30 days but early needling occurred for total lack of alternative access. No problems were encountered in these early-needled grafts. The primary reason for delayed access in nine patients (>60 days) was limb oedema, graft needling not required or central stenosis requiring prior treatment. In this series, varying degrees of oedema or limb swelling was reported in seven grafts (two with known central vein stenosis) all seven patients with early oedema went on to undergo haemodialysis.

In these two patients fusiform dilatation (<50% increase in diameter) was seen only at needling sites and no graft was lost or revised for aneurysmal dilatation. Both grafts continue to be needled at last reported follow-up with no further changes in graft size (930 and 602 days). Distal revision with a second piece of SGVG 100 was performed in five grafts. One graft developed an anastomotic pseudoaneurysm at the arterial anastomosis and underwent an uneventful localised surgical repair. A summary of all reported complications is listed in Table 2. There were four mortalities reported in the series due to gastrointestinal bleeding at 369 days, chronic renal failure at 486 days, anoxic brain damage at 544 days and sepsis at 601 days. The deaths were not related to the SGVG 100, and all grafts were reported as patent at the time of death.

Table 2. Reported graft complications
Events*SGVG 100 grafts# of grafts w/eventLinearised rate (events/year)
Occlusion19120.75
Stenosis32†141.26
Dilatation220.08
Infection220.08
Haematoma220.08
Pseudoaneurysm110.04
Bleeding complication110.04
Steal000.00
Rupture000.00
Oedema/swelling7‡70.28

*More than one complication may be reported for a single graft at follow-up.

†Six in the native vein.

‡Two with known central vein stenosis.

Two SGVG 100 grafts in the series developed complications related to infection. In the first, a patient developed an infection on day 587, which was initially treated with antibiotics. The graft underwent revision on day 604 to replace the infected needling site with a second SGVG 100. The old section continued to be needled and the graft remains patent at last follow-up. The second infection in a graft placed in the upper thigh, developed on day 122 and underwent a revision with a short replacement section of new SGVG 100. The surgery was initially successful, but the graft thrombosed 4 days postoperative after insertion of a tunnelled line in the ipsi-lateral common femoral vein and could not be saved via thrombectomy.

A total of 19 occlusions were reported in 12 SGVG 100 grafts (Table 2). The mean time to occlusion was 215±141 days (range 34–604 days). Surgery to re-establish patency was successful in 14 of 18 attempts. One patient refused exploration after the SGVG 100 graft clotted due to a hypotensive episode.

As shown in Table 3, in 19 of the 25 cases a total of 49 procedures have been required in an attempt to maintain or re-establish patency. Hence, the intervention rate to maintain or re-establish patency was 1.94 per graft year. Thirty interventional radiology procedures in 14 SGVG 100 grafts were performed to treat luminal stenosis (Table 3). Angioplasty was the primary treatment technique (n=30) with patency re-established in all 30 procedures (four of 14 grafts accounted for 17 procedures).

Table 3. Reported treatment procedures
Treatment type*SGVG 100 grafts (n=19)
Angioplasty30
Thrombolysis0
Thrombectomy12†
Graft explant or ligated1
Graft revision‡6

*More than one treatment may be reported for a single graft at follow-up.

†Concomitant angioplasty and stent placement in a single graft.

‡Five grafts with a second piece of bovine graft inserted.

Primary patency, assisted primary patency, secondary patency, and freedom from infection were 29, 49, 81, and 95% at 1 year, respectively (Table 4, Fig. 1). The event rates are shown in Table 5, demonstrating the incidence of thrombosis to be 0.75 events per graft year with an average of 0.76 occlusions per graft.

Table 4. Actuarial outcome table of performance variables at 1 year
Performance variable# of events% Event freeStandard error (%)Lower 95% CI (%)Upper 95% CI (%)
Primary patency182910949
Primary assisted patency1245112367
Secondary patency68196399
Mortality41000100100
Infection295588100

Confidence limits (95%) were calculated as 1.96×SE±event (%).

Table 5. Linearised event rates
EventNumber of eventsLinearised rate (events/year)
Infection20.08
Thrombosis190.75
Total Interventions491.94
Abandonment (infection)*10.04
Abandonment (thrombosis)40.16
Abandonment (all)60.24

*Infection+occlusion in a single graft.

Back to Article Outline

4. Discussion 

All those responsible for vascular access surgery will recognise a patient group in which conventional vascular access has failed or is not possible. Previous biological grafts have been used for this type of patient group. These older xenografts were chemically fixed to destroy xeno-antigenicity. However, this makes them less biocompatible. There is also a tendency to aneurysm formation, late aldehyde initiated calcification and thrombosis.23, 24 Graft handling and suturing of the SGVG 100 was found to be straightforward. It is advisable to avoid clamping the graft, excessive traction or handling of the collagen scaffold until it repopulates. With the graft studied there were no early thromboses or infections (30 day) showing that the graft is easily inserted resistant to kinking and clotting and the patient does not require anticoagulation. Needling for dialysis was delayed to 30 days after insertion. This was based on the assumption that the graft will become stronger with time as host cells incorporate into a collagen matrix.25, 26 There is evidence from dog work21 and from grafts in this series (data not shown) that capillaries, myocytes, fibroblasts, neo collagen and endothelium do indeed appear.27 In several patients needling was performed early (14 days) after alternative access was lost. This did not prove to be a problem and it is possible that the graft could be needled earlier but we have no experience of this.

The study group included patients with absence of native vein, failed prior grafts, active infection, obesity and central vein stenosis. The patient demographics, time on dialysis and prior surgical attempts at access reflect this unfavourable patient study group (Table 1). As a consequence it is difficult to compare the results from this series with published data using for example PTFE. Furthermore, we know that the threshold for the use of PTFE is much lower in the USA where most of the large studies of PTFE have been performed.28 In a large survey of PTFE in the USA, primary patency rate was 38% at 1 year and 25% at 2 years.29 PTFE was found to have a 72% secondary patency at 12 months when 70% of PTFE grafts were the first graft procedure performed.29 By comparison secondary patency for SGVG in this series was 81%, with only 4% (one of 25) of grafts places as the first access procedure. This secondary patency meets the National Kidney Foundation-Dialysis Outcomes Initiative goal of 70% at 12 months.4

One year primary patency without intervention was 29% but improved to 45% when considering grafts that were treated by angioplasty or revision without thrombosis. It had been hoped that the improved biomechanical properties of the graft might ameliorate the venous outflow stenosis seen with synthetic grafts. However, this has not been the case in these high-risk patients. Stenoses tended to occur in the distal end of the graft and not in the native vein unlike with PTFE and this made it easier to salvage. Stenoses when resected were proliferative in nature on histology (data not shown) and those treated radiologically responded readily to angioplasty.30 As a result we now assess the grafts regularly by thermo-dilution flow monitor initially every 3 months and perform duplex ultrasound on any graft showing decreasing flow characteristics or rising venous pressures. It would appear for the SGVG graft as with PTFE,31 that a high level of monitoring and intervention will still be required.

Infection rates have been low and it would appear that the rate is lower than historical reports on PTFE (circa 10%).5, 7, 8, 9 This may well be the greatest asset for the graft. It could be postulated that the graft would have the ability to resist and remove infection. This would be supported by the low infection rates in the dog,21 in this study and when cryopreserved non cross-linked tissue was placed directly in the infected field.10, 11, 12

In conclusion, the SGVG 100® graft may be advantageous to other biological grafts as it has wall compliance properties that may limit myo intimal hyperplasia, has a substantial wall thickness to avoid dilatation, ability to repair wall damage from needling, and ability to remove infection form needle tracts.27 Our initial experience of the SGVG 100® shows that it was a safe and promising tool in these difficult patients. Results should improve further as less challenging patients are studied. There would now be a strong case for comparison with PTFE in a randomised trial but perhaps still concentrating on tertiary vascular access patients.

Back to Article Outline

Acknowledgements 

The manufactures of the bovine ureteric graft have provided the grafts at the full market price (SGVG 100, CryoLife, Inc., Kennesaw, GA, USA). Cryolife made a grant to the Department Endowment fund for research to contribute to the staff time for data collection and monitoring. An online database was set up by the Cryolife to facilitate paperless prospective data collection. All data was collected, inputted and statistics verified by the authors.

Back to Article Outline

References 

  1. US Renal Data System (USRDS). Annual Data Report: Atlas of end Stage renal disease in the United States. Bethesda, MD: National Institute of Health, National Institute of Diabetes and Digestive and Kidney Diseases; 2004;
  2. US Renal Data System (USRDS). Annual Data Report: Atlas of end stage renal disease in the United States. Bethesda, MD: National Institute of Health, National Institute of Diabetes and Digestive and Kidney Diseases; 2003;
  3. National Kidney Foundation. K/DOQI clinical practice guidelines for vascular access. Am J Kidney Dis. 2001;37(Suppl 1):S137–S181
  4. Chazan JA, London MR, Pono LM. Long-term survival of vascular accesses in a large chronic hemodialysis population. Nephron. 1995;69(3):228–233
  5. Cinat ME, Hopkins J, Wilson SE. A prospective evaluation of PTFE graft patency and surveillance techniques in hemodialysis access. Ann Vasc Surg. 1999;13(2):191–198
  6. Feldman HI, Kobrin S, Wasserstein A. Hemodialysis vascular access morbidity. J Am Soc Nephrol. 1996;7(4):523–535
  7. Hodges TC, Fillinger MF, Zwolak RM, Walsh DB, Bech F, Cronenwett JL. Longitudinal comparison of dialysis access methods: Risk factors for failure. J Vasc Surg. 1997;26(6):1009–1019
  8. Matsuura JH, Rosenthal D, Clark M, Shuler FW, Kirby L, Shotwell M, et al. Transposed basilic vein versus polytetrafluorethylene for brachial-axillary arteriovenous fistulas. Am J Surg. 1998;176(2):219–221
  9. Taylor SM, Eaves GL, Weatherford DA, McAlhany JC, Russell HE, Langan EM. Results and complications of arteriovenous access dialysis grafts in the lower extremity: A five year review. Am Surg. 1996;62(3):188–191
  10. Matsuura JH, Johansen KH, Rosenthal D, Clark MD, Clarke KA, Kirby LB. Cryopreserved femoral vein grafts for difficult hemodialysis access. Ann Vasc Surg. 2000;14(1):50–55
  11. Surowiec S. The use of cryopreserved femoral vein for salvage of infected arteriovenous grafts: Early results. Baltimore, MD: Association of VA surgeons; 1998;
  12. Matsuura J. Cryopreserved human femoral vein: A new option for infected access grafts. Contemporary dialysis and nephrology. p. 30–32 1999;
  13. Sabanayagam P, Schwartz AB, Soricelli RR, Lyons P, Chinitz J. A comparative study of 402 bovine heterografts and 225 reinforced expanded PTFE grafts as AVF in the ESRD patient. Trans Am Soc Artif Intern Organs. 1980;26:88–92
  14. Enzler MA, Rajmon T, Lachat M, Largiader F. Long-term function of vascular access for hemodialysis. Clin Transplant. 1996;10(6 Pt 1):511–515
  15. Brems J, Castaneda M, Garvin PJ. A five-year experience with the bovine heterograft for vascular access. Arch Surg. 1986;121(8):941–944
  16. Bacchini G. PTFE versus mesenteric bovine vein as vascular accesses in hemodialysis patients: Preliminary results of a randomized trial. London, UK: International Congress of the Vascular Access Society; 2001;
  17. Ratto GB, Romano P, Truini M, Frascio M, Rovida S, Badini A, et al. Glutaraldehyde-tanned mandril-grown grafts as venous substitutes. J Thorac Cardiovasc Surg. 1991;102(3):440–447
  18. Ratto G. Reply to the Editor regarding J Thorac Cardiovasc Surg 1991. J Thorac Cardiovasc Surg. 1994;107:949
  19. Widmer MK, Aregger F, Stauffer E, Savolainen H, Heller G, Hakki H, et al. Intermediate outcome and risk factor assessment of bovine vascular heterografts used as AV-fistulas for hemodialysis access. Eur J Vasc Endovasc Surg. 2004;27(6):660–665
  20. Clarke DR, Lust RM, Sun YS, Black KS, Ollerenshaw JD. Transformation of nonvascular acellular tissue matrices into durable vascular conduits. Ann Thorac Surg. 2001;71(5 Suppl 1):S433–S436
  21. Matsuura JH, Black KS, Levitt AB, Rosenthal D, Wellons ED, Fallon MT, et al. Cellular remodeling of depopulated bovine ureter used as an arteriovenous graft in the canine model. J Am Coll Surg. 2004;198(5):778–783
  22. Sidawy AN, Gray R, Besarab A, Henry M, Ascher E, Silva M, et al. Recommended standards for reports dealing with arteriovenous hemodialysis accesses. J Vasc Surg. 2002;35(3):603–610
  23. Schmitz-Rixen T, Megerman J, Anderson JM, Warnock DF, L'Italien GJ, Erasmi H, et al. Longterm study of a compliant biological vascular graft. Eur J Vasc Surg. 1991;5(2):149–158
  24. Koch G, Gutschi S, Pascher O, Fruhwirth H, Glanzer H. Analysis of 274 omniflow vascular prostheses implanted over an eight-year period. Aust N Z J Surg. 1997;67(9):637–639
  25. Raeder RH, Badylak SF, Sheehan C, Kallakury B, Metzger DW. Natural anti-galactose alpha1,3 galactose antibodies delay, but do not prevent the acceptance of extracellular matrix xenografts. Transpl Immunol. 2002;10(1):15–24
  26. Badylak S, Kokini K, Tullius B, Whitson B. Strength over time of a resorbable bioscaffold for body wall repair in a dog model. J Surg Res. 2001;99(2):282–287
  27. Darby C, Cornall A. Clinical outcomes of a depopulated bovine ureter (Synergraft Vascular Graft model 100) used as an arteriovenous access graft. Advances in tissue engineering and biology of heart valves, Florence, Italy, 2004.
  28. Rayner HC, Pisoni RL, Gillespie BW, Goodkin DA, Akiba T, Akizawa T, et al. Creation, cannulation and survival of arteriovenous fistulae: Data from the dialysis outcomes and practice patterns study. Kidney Int. 2003;63(1):323–330
  29. Gibson KD, Gillen DL, Caps MT, Kohler TR, Sherrard DJ, Stehman-Breen CO. 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(4):694–700
  30. Warakaulle D, Cornall A, Uberoi R, Darby C. Vascular bovine ureteric grafts (VBUGS) as a novel conduit in fashioning haemodialysis fistulae. Chicago, IL: RSNA; 2004;
  31. Ruddle A, Lear P, Mitchell D. The morbidity of secondary vascular access. A life time of intervention. Eur J Vasc Endovasc Surg. 1999;18(1):30–34

PII: S1078-5884(05)00463-6

doi:10.1016/j.ejvs.2005.07.006

European Journal of Vascular & Endovascular Surgery
Volume 31, Issue 2 , Pages 181-186, February 2006

Article Tools

* Image Usage & Resolution